RR439B-Ethnic minorities and attainment the effects of poverty annex.pdf

Ethnicity, deprivation
and educational
achievement at age 16
in England: trends over
time
Annex to compendium of evidence on
ethnic minority resilience to the effects
of deprivation on attainment
June 2015
Professor Steve Strand - University of
Oxford
Contents
List of figures
5
List of tables
6
Executive Summary
7
Introduction
7
Achievement Gaps
7
1991-2006: Youth Cohort Study
7
2003-2013: Five or more A*-C grades (5AC)
8
2004-2013: Five or more A*-C grades including English and mathematics
(5EM)
8
Capped (Best 8) point score
9
Relative size of equity gaps (ethnicity, gender and entitlement to FSM)
9
Ethnic achievement gaps by sub-groups: FSM and Gender
10
Adjusting ethnic achievement gaps to account for socio-economic
disadvantage
10
Ethnic achievement gaps at primary school
11
Introduction
12
Methodology
14
Data sources
14
Focal set of ethnic minority groups
14
Percentages and Odds Ratios
15
The Ethnic minority population age 5-16 in England
17
National data and trends 2003 - 2013
17
Ethnic minority population by region
18
Ethnic minority population by Local Authority (LA)
19
2
Ethnic trends in educational attainment
24
1991-2006: The Youth Cohort Study
24
2003-2013: The National Pupil Database: 5+ GCSE A*-C grades or equivalent
(5AC)
25
2004-2013: 5+ GCSE A*-C or equivalent including English & mathematics
(5EM)
29
Introduction
29
Contrasting 5AC and 5EM
30
Ethnic achievement gaps on 5EM 2004-2013
31
Capped (Best 8) points scores
33
Overview across the three indicators
35
Trends in the relative size of ethnic, gender and FSM gaps
37
Achievement by combinations of ethnic group, entitlement to FSM and
gender
40
Pupils not entitled to FSM
41
Pupils entitled to FSM
44
Check against Best 8 score
48
Adjusting ethnic achievement gaps for socio-economic factors
Ethnic achievement gaps at primary school
50
53
End of Key Stage 2 (age 11)
54
End of Foundation Stage (age 5)
55
Other educational outcomes
56
References
57
Appendices
61
Appendix 1: DfE Statistical First Releases (SFR) relating to age 16 attainment 62
Appendix 2: Pupil numbers at the end of key stage 4 by ethnic group,
entitlement to FSM and gender for 2013
3
63
Appendix 3: Notes on the Local Authority maps for percentage ethnic
minority students 2003 & 2013
64
Appendix 4: GCSE 5AC by ethnic group and gender: 2003-2013
65
Appendix 5: GCSE 5EM for pupils not entitled to FSM by ethnic group and
gender: 2004-2013
66
Appendix 6: GCSE 5EM for pupils entitled to FSM by ethnic group and
gender: 2004-2013
67
Appendix 7: Key stage 2 achievement by ethnic group, gender and
entitlement to FSM: 2013
68
Appendix 8: Foundation Stage Profile achievement by ethnic group, gender
and entitlement to FSM
69
Appendix 9: Special educational needs (SEN) and fixed term exclusions
(FTE)
70
Special educational needs (SEN)
70
Exclusions from school
71
4
List of figures
Figure 1: Percentage ethnic minority students by Local Authority: England 2003
21
Figure 2: Percentage ethnic minority students by Local Authority: England 2013
22
Figure 3: Percentage of 16 year olds achieving 5 or more GCSEs at grades A*-C: YCS
1991-2006.
24
Figure 4: Percentage 5+ GCSE A*-C or equivalent by ethnic group and gender: 20032013
27
Figure 5: Percentage of pupils achieving 5+ A*-C GCSE or equivalent by ethnic group:
2003-2013
28
Figure 6: Percentage of all students achieving 5EM by ethnic group: 2004-2013
32
Figure 7: Best 8 points score by ethnic group: Selected years 2004-2013
35
Figure 8: Trends in ethnic, gender and FSM achievement gaps at age 16 over time:
2004-2013
39
Figure 9: Cohen's D effect size for gender, ethnic and FSM Achievement gaps in 2004
and 2013
40
Figure 10: GCSE 5EM for pupils not entitled to FSM by ethnic group and gender: 20042013
42
Figure 11: GCSE 5EM for all pupils not entitled to FSM by ethnic group: 2004-2013 43
Figure 12: GCSE 5EM for pupils entitled to FSM by ethnic group and gender: 20042013
45
Figure 13: 5+ A*-C GCSE (Inc. English & Maths) for all pupils entitled to FSM by ethnic
group: 2004-2013
47
Figure 14: Best 8 points score by ethnic group and entitlement to FSM
49
Figure 15: Percentage of students age 5-16 entitled to a FSM: England 2011
51
Figure 16: Percentage of students in each of five score bands based on IDACI:
England 2011
51
Figure 17: GCSE results at age 16 adjusted for Socio-economic Status (SES)
52
Figure 18: Age 16 5EM results by ethnic group and FSM: England 2013
53
Figure 19: Age 11 results by ethnic group and FSM: England 2013
5
54
Figure 20: Age 5 results by ethnic group and FSM: England 2013
55
Figure 21: Unadjusted and adjusted ethnic group Odds Ratios for identification of
BESD: January 2011
71
Figure 22: Number of exclusions during secondary school (Y7-Y11)
72
List of tables
Table 1: Proportion of ethnic minority students: England 2003 and 2013
18
Table 2: Percentage of students in each ethnic group by region, England 2013
19
Table 3: Number and percentage of ethnic minority pupils: Highest and lowest LAs:
2013.
23
Table 4: Ethnic group Odds Ratios for the 2004 cohort for both 5EM and 5AC
measures.
30
Table 5: Percentage of all students achieving 5EM by ethnic group: 2004-2013
31
Table 6: Best 8 points score by ethnic group: Selected years 2004-2013
34
Table 7: Trends in ethnic, gender and FSM achievement gaps at age 16 over time:
2004-2013
37
Table 8: GCSE 5EM for all pupils not entitled to FSM by ethnic group: 2004-2013
43
Table 9: 5+ A*-C GCSE (Inc. English & Maths) for all pupils entitled to FSM by ethnic
group: 2004-2013
46
Table 10: Best 8 points score by ethnic group and entitlement to FSM
6
48
Executive Summary
Introduction
This report was commissioned by the Department for Education (DfE) to:
•
Establish trends in ethnic achievement over the last 10 years (and longer where
possible) focussed on age 16 outcomes; and
•
Look at the level of ethnic group, but also where possible further disaggregate
the data by entitlement to FSM and gender to determine whether particular
patterns of change have occurred within sub-groups.
The report is essentially descriptive. It is intended to support a separate literature
review commissioned by the DFE to explore reasons underlying ethnic minority
achievement gaps and possible changes over time.
Achievement Gaps
1991-2006: Youth Cohort Study
Early trends (1991-2006) are explored using the Youth Cohort Study (YCS) which gives
the only nationally representative data on ethnic minority achievement collected from
the early 1990's. The results indicate that all ethnic minority groups made significant
progress in the period 1991-2006, though the YCS does not allow breakdowns within
the ‘White’ and ‘Black’ groups.
•
Indian pupils moved from an equivalent position to White students in 1991 to be
significantly ahead in 2006, and Other Asian pupils further increased their lead.
•
Bangladeshi pupils made the greatest progress. In 1991 just 14% of
Bangladeshi pupils achieved the benchmark of 5+ GCSE A*-C grades (5AC)
compared with 37% of White pupils. By contrast, in 2006 57% of Bangladeshi
pupils achieved this benchmark against 58% of White pupils.
•
The Pakistani and Black achievement gaps narrowed substantially. Both
groups doubled the proportion achieving 5AC, from 26% to 52% for Pakistani
pupils and from 23% to 50% for Black pupils. Most of the improvement,
particularly for Black pupils, occurred between 2003 and 2006.
7
2003-2013: Five or more A*-C grades (5AC)
From 2003 onwards data on the ethnicity of all students in England was captured
through the Annual School Census and matched to age 16 achievement data. Analysis
of this data reveals:
•
Indian and Chinese pupils have consistently outperformed the White British
average. For both groups the odds of achieving 5AC are currently over twice as
high as for White British pupils.
•
In 2003 the odds for Black Caribbean pupils achieving 5AC were less than half
the odds for White British pupils (32.9% vs. 51.3%, OR 1=0.47), and for Mixed
White & Black Caribbean and Black African pupils only two-thirds the odds for
White British pupils (39.9% with OR=0.63 and 40.7% with OR=0.67
respectively). In 2013 Black African pupils achieve slightly above the White
British average (OR=1.11) and both Black Caribbean and Mixed White & Black
Caribbean students have closed the gap with White British students, from 18
percentage points in 2003 to just 2 percentage points (80.4% vs. 82.7%) in
2013, although a small gap remains (OR= 0.86 for both minority groups).
•
A very large improvement for Bangladeshi students occurred in the years from
1991 to 2003, but in 2003 they were still below the White British average
(OR=0.79). However, by 2013 Bangladeshi students are achieving above the
White British average (OR=1.19). This is despite Bangladeshi students being
one of the most socio-economically deprived communities in England.
•
Pakistani pupils have also improved substantially. In 2003 their odds of
achieving 5AC were only two thirds the odds for White British students (41.5%
vs. 51.3%, OR=0.67). However, they closed the gap steadily and achieved parity
with White British students in 2011 and currently their odds of achieving 5AC do
not significantly differ from White British students (OR=1.07).
2004-2013: Five or more A*-C grades including English and
mathematics (5EM)
The 5AC indicator has limited contemporary currency since it was superseded in 2006
by a new indicator of 5 or more GCSE A*-C grades or equivalent including English and
mathematics (5EM). In 2004, the proportion of students achieving 5AC was 51.9%
while the proportion achieving 5EM was only 40.9%, a difference of 11 percentage
1
The Odds Ratio (OR) compares the odds of achieving success on a particular measure (e.g. 5EM) for an ethnic
minority group against the odds for a reference group, here White British students. If the odds of achieving 5EM
are the same for the ethnic minority group as for White British students the OR=1, if the odds for the ethnic
minority are lower than for White British students then the OR<1 and if they are higher the OR>1.
8
points. The impact of the move to 5EM was greatest for ethnic minorities and the ethnic
achievement gaps for 5EM were larger than for 5AC. However, over the subsequent 10
years from 2004 to 2013:
•
Indian and Chinese students continue to have the highest achievement and in
2013 their odds of achieving 5EM are twice the odds for White British students.
•
Bangladeshi students substantially increased in achievement from well below the
White British average in 2004 (OR=0.68) to above the White British average in 2013
(OR=1.16).
•
Black African students have also show substantial increases and there is no
longer any gap relative to White British, the OR changing from OR=0.65 in 2004
to OR=1.03 in 2013.
•
There has been a significant closing of the gap for Black Caribbean, Mixed White
& Black Caribbean and Pakistani students against White British students, though
gaps do remain. For example the odds for Black Caribbean students achieving 5EM
were less than half the odds for white British in 2004 (OR=0.43) but in 2013 this
had narrowed to OR= 0.75. For Mixed White & Black Caribbean students there
has been a change in OR from 0.53 to 0.79 and for Pakistani students from 0.64
to 0.81. However all three of these ethnic groups are achieving less well than
White British since their odds of achieving the threshold are still around 20%
lower than for White British students.
•
Only one ethnic group has seen performance decline relative to White British
over the period. White other groups have moved from above the White British
average in 2004-2006 to below the White British average from 2008 onwards
and in 2013 the OR=0.81 is similar to that for Black Caribbean, Mixed White &
Black Caribbean and Pakistani students. The decrease is likely to reflect
demographic change, particularly increased entry from Eastern Europe.
Capped (Best 8) point score
The analyses were repeated for a continuous measure, capped or Best 8 points score
and the trends described above were essentially the same for this measure.
Relative size of equity gaps (ethnicity, gender and
entitlement to FSM)
We have seen above that ethnic achievement gaps have reduced substantially over the
last 20 years in absolute terms. They have also reduced substantially relative to the
size of other equity gaps, such as the gender gap and the socio-economic
disadvantage gap, as indicated by entitlement to a Free School Meal (FSM). In 2004
9
the ethnic gap on 5EM was 18.1 percentage points, more than twice as large as the
gender gap (7.7% points) and half the size of the FSM gap (28% points). However, by
2013 this had changed substantially. Now the ethnic gap (7.2% points) is actually
smaller than the gender gap (10.1% points), and only about one-quarter of the size of
the FSM gap (26.7% points). Similar proportional changes were seen for Best8 points
score. This further attests to the scale of the reduction in the ethnic achievement gap.
Ethnic achievement gaps by sub-groups: FSM and Gender
The most striking feature of sub-groups is that among pupils entitled to FSM all
ethnic minority groups achieve greater success than White British pupils. In
2013, relative to White British students, the odds for Chinese pupils achieving 5EM
were 6.9 times greater, for Indian students 3.4 times greater, for Bangladeshi 3.0 times,
for White other 1.6 times, for Black Caribbean 1.5 times and even for the lowest
achieving minority group, Mixed White & Black Caribbean students, the odds of
achieving 5EM were 1.26 times (26%) higher than for White British FSM students.
Pupils entitled to FSM from ethnic minority groups pulled even further ahead of
White British FSM students between 2004 and 2013. Between 2004 and 2013 the
OR for Mixed White and Black Caribbean students increased from OR=0.97 to
OR=1.26, for Black Caribbean students it increased from OR=0.98 to OR=1.53, for
Black African students from OR=1.44 to OR=2.22 and for Bangladeshi students from
OR=2.52 to OR=3.0. This trend was not evident in Best 8 score, so may reflect an
improved likelihood of achieving A*-C grades in English and mathematics in particular.
In terms of 5EM, White British FSM students are the lowest achieving ethnic group and
the achievement gap with most ethnic minority groups appears to have increased 20042013.
Adjusting ethnic achievement gaps to account for socioeconomic disadvantage
We have seen that while ethnic achievement gaps have reduced remarkably over the
last 10 years some (smaller) gaps do still exist for Black Caribbean, Mixed White &
Black Caribbean and Pakistani students. However, the above data take no account of
the fact that levels of socio-economic disadvantage are considerably higher among
ethnic minority groups than the White British group. Studies that use regressions
methods to control for socio-economic status (SES) suggest almost all ethnic minority
groups achieve higher exam success at age 16 than White British students of the same
SES. For example Strand (2014a) reports that all ethnic minority groups achieve at
least as well as, and frequently substantially better than, the White British students at
age 16, with the single exception of middle and high SES Black Caribbean boys.
10
Ethnic achievement gaps at primary school
The very low achievement of White British students entitled to FSM is not just a feature
of educational achievement at age 16. Analysis of end of Key Stage 2 results at age 11
indicates that White British FSM pupils (58%) and White Other FSM pupils (56%) are
the two lowest achieving groups at age 11. Equally an analysis of the Early Years
Foundation Stage (EYFS) Profile at age 5 indicates that White British pupils entitled to
FSM (35%), along with White Other FSM (34%) and Pakistani FSM (34%) pupils are
the three lowest achieving groups.
While White Other and Pakistani FSM pupils achieve as poorly as White British FSM
pupils at age 5, many pupils in these groups have relatively limited experience with
English language (on starting school, or are recent arrivals in England with low fluency
in the English language (see Strand, Malmberg & Hall, 2015). When they subsequently
acquire this fluency they make rapid educational progress while White British pupils do
not. For example in their analysis of the National Pupil Database, Strand et al. (2015)
show that the EAL gap is largest in the EYFS at age 5 (Good level of development
OR=0.67) but more or less disappears by age 16 (5EM OR=0.90).
These results indicate a comprehensive account of the drivers of the low educational
achievement of White British pupils on FSM at age 16 must include consideration of
factors operating in the first five years of the child's life.
11
Introduction
This report was commissioned by the Department for Education (DfE) to:
•
Establish robust trends in ethnic achievement over the last 10 years (and longer
where possible) focussed on age 16 outcomes; and
•
Look at the level of ethnic group, but also where possible further disaggregate
the data by FSM status and gender to determine whether particular patterns of
change have occurred within sub-groups.
The report is essentially descriptive. It is intended to support a separate literature
review commissioned by the DfE to explore reasons underlying ethnic minority
achievement gaps and possible changes over time.
•
It starts by outlining demographic details on the size of the ethnic minority school
population age 5-16 in England, the distribution across English regions and
Local Authorities (LA), and the growth in the ethnic minority population 20032013.
•
Next data are presented from the Youth Cohort Study (YCS) which give the only
nationally representative data on ethnic minority achievement collected from the
early 1990's. The report then moves to a discussion of national data on ethnicity
and achievement collected annually from all pupils in the school census of 2003
onwards. We consider trends in three indicators; the proportion of students
achieving 5 or more GCSE A*-C grades or equivalent (5AC); the proportion of
students achieving 5 or more GCSE A*-C grades or equivalent including English
and maths (5EM); and the capped or ‘Best 8’ points score.
•
Next the report considers not only absolute changes in the achievement of
different ethnic minority groups but also the relative size of the ethnic, gender
and Free School Meal (FSM) achievement gaps and how these have changed
over time.
•
After considering data for each ethnic minority group as a whole, we look in
detail at the breakdown on the 5EM measure by the combination of ethnic group,
entitlement to FSM and gender, to explore trends by sub-groups.
•
We consider achievement gaps by ethnicity and FSM at earlier ages during
primary school, looking at achievement in national tests at age 11 and at the
Foundation Stage Profile at age 5. These data highlight that achievement gaps
are apparent from the very early stages of schooling.
12
•
Data on two other educational outcomes, the identification of Special
Educational Needs and fixed term exclusions from school, are included as an
Appendix.
13
Methodology
Data sources
Data has in the main been drawn from DfE Statistical First Releases (SFR), although
some statistics have been calculated by the author directly from National Pupil
Database (NPD) files. Appendix 1 lists the SFRs drawn on in collating data on
achievement up to age 16.
Focal set of ethnic minority groups
National data is collected for eighteen ethnic groups (plus refused/unclassified) but
presentation of this full set of codes in figures can be difficult to read. Also, even though
we are dealing with data for the whole of England, in any particular year group the
numbers can be quite small, especially for the Gypsy Roma and Irish Traveller groups
which represent just a few hundred pupils (approx. 0.15% of the population). When data
are further broken down by entitlement to FSM and gender some ethnic groups also
have quite small sample sizes. In the graphs in this report the following nine ‘focal’
ethnic groups will be displayed:
•
White British
•
White Other
•
Indian
•
Pakistani
•
Bangladeshi
•
Chinese
•
Black African
•
Black Caribbean
•
Mixed White & Black Caribbean
The rationale for highlighting these nine ‘focal’ groups is given below. Where
numbers are quoted these represent numbers of students in Year 11 in 2013 (see
Appendix 2).
14
•
White British represent the majority of students in England at just under threequarters (73%) of the age 5-16 school population in 2013.
•
White Other is now the largest ethnic minority group in England and one of the
fastest growing (n=19,265 in 2013 representing 3.4% of all Y11 students). It
consists of pupils from a wide range of nationalities and speaking a wide range
of languages, but predominantly from Europe including Polish (21%), Turkish
(7%), Portuguese (5%), Albanian (4%) and Lithuanian (4%) (See Strand et al,
2015).
•
Indian, Pakistani and Bangladeshi groups have a long history of residence as
communities in England and it is important to report their results separately given
the strong contrasts in educational achievement between these groups.
•
Chinese are the smallest ethnic group included (n=2,257 in 2013) but are
included because considerable interest has focused on the strong educational
performance of this group.
•
Black African students are the largest Black group (n=16,201) twice as
numerous as Black Caribbean. They are also one of the two fastest growing
groups, along with White Other groups.
•
Black Caribbean students have long been a focus for concern over low
educational achievement and this merits continued monitoring.
•
Mixed White and Black Caribbean are the largest group of students of mixed
heritage accounting for around one-third (34%) of all mixed heritage students.
There are also now as many Mixed White & Black Caribbean students (n=7,333)
as there are Black Caribbean students (n=8,158).
This focus in no way implies that the achievement of other ethnic groups is not of
interest and the results of all ethnic groups are included in the report tables. The focus
on the nine groups identified above allows the calculation of additional statistics and
graphical presentations to be kept to manageable levels.
Percentages and Odds Ratios
Given there are over 570,000 students in the typical year group in England, statistical
significance may be a poor guide to educational significance due to the huge sample
size. In gauging the size of the gaps in educational achievement here we present the
absolute gaps between ethnic groups in terms of percentage points or in terms of mean
scores. We also calculate standardised effect size measures, such as the Odds Ratio
(OR) for dichotomous measures (5AC and 5EM) and Cohen's D for continuous
measures (Best 8 points score). The Odds Ratio (OR) compares the odds of achieving
15
success on a particular measure (e.g. 5EM) for an ethnic minority group against the
odds for a reference group, here White British students. If the odds of achieving 5EM
are the same for the ethnic minority group as for White British students the OR=1, if the
odds for the ethnic minority are lower than for White British students then the OR<1 and
if they are higher the OR>1. There is no absolute level at which an OR is educationally
significant or substantial. However, as a rough rule we might consider that if the odds of
success for an ethnic minority group were 25% higher (OR=1.25) or 25% lower
(OR=0.75) than for White British students this would be a difference worthy of
recognition.
16
The Ethnic minority population age 5-16 in England
National data and trends 2003 - 2013
In 2013 1.77M students, over one-quarter (26.6%) of the age 5-16 maintained school
population in England, were from ethnic minority groups. Table 1 presents the
proportion of students in each ethnic minority group and contrasts the 2013 and 2003
data. The selection of base year is made because although national data were first
collected in 2002 a new ethnic classification system was introduced in 2003 which was
not compatible with the previous year. Specifically (a) the breakdown of the "White"
category into White British, White Irish, Travellers, Gypsy-Roma and White Other
groups, and (b) the introduction of four mixed-heritage categories, make the data
incompatible with that collected in 2002.
Comparing across 2003 and 2013, the overall proportion of White British students has
decreased from 83.2% to 73.4% of the school population, or conversely the ethnic
minority population has increased from 16.8% in 2003 to 26.6% in 2013.
The two ethnic groups with the largest absolute increases are White Other Background
(from 2.1% to 4.3% of the population) and Black African (from 1.7% to 3.3% of the
population). Black African students are now the largest of the Black ethnic groups. Also
notable is that the number of Mixed White & Black Caribbean students is now the same
as the number of Black Caribbean students (both 1.4% of the population).
17
Table 1: Proportion of ethnic minority students: England 2003 and 2013
Ethnic group
White
White British
Irish
Traveller of Irish heritage
Gypsy/ Roma
Any other White
background
Mixed
White and Black Caribbean
White and Black African
White and Asian
Any other Mixed
background
Asian
Indian
Pakistani
Bangladeshi
Chinese
Any other Asian
background
Black
Black Caribbean
Black African
Any other Black
background
Any other ethnic group
Classified
Unclassified
Minority Ethnic Pupils
All pupils
2003
N
%
5,590,100 85.9
5,418,900 83.2
26,500 0.4
3,800 0.1
6,000 0.1
2013
N
5,207,830
4,877,300
21,800
4,555
16,735
%
78.3
73.4
0.3
0.1
0.3
Change
% points
-7.5
-9.9
-0.1
0.0
0.2
134,900
169,000
60,700
15,000
33,300
2.1
2.6
0.9
0.2
0.5
287,435
306,890
92,505
36,730
68,605
4.3
4.6
1.4
0.6
1.0
2.3
2.0
0.5
0.3
0.5
60,000
440,600
153,800
175,200
70,300
22,800
0.9
6.8
2.4
2.7
1.1
0.4
109,060
678,680
175,035
262,535
107,320
107,815
1.6
10.2
2.6
3.9
1.6
0.4
0.7
3.4
0.3
1.3
0.5
0.0
41,300
233,000
97,300
108,400
0.6
3.6
1.5
1.7
25,975
353,915
90,455
220,785
1.6
5.3
1.4
3.3
1.0
1.7
-0.1
1.7
27,300 0.4
54,300 0.8
6,509,800 100
272,600 4.0
1,930,220 16.8
6,782,400
42,675
100,860
6,648,195
64,450
1,770,895
6,712,645
0.6
1.5
100
1.0
26.6
0.2
0.7
0.0
-3.1
9.9
Note: Based on students of compulsory school age (5-16 years). Percentages exclude unclassified (4% of
all students in 2003 and 1% in 2013). Data drawn from DFE SFR 09/2003 and DFE SFR 21/2013
Ethnic minority population by region
There are large regional variations in the proportion of ethnic minority students. Data by
region from the latest school census are presented in the table below.
18
Table 2: Percentage of students in each ethnic group by region, England 2013
White British
England
North
East
South
West
North
West
South
East
East Yorkshire
East
England & Humber Midlands
West
Midlands
Outer
London
Inner
London
Greater
London
73.3
91.6
89.4
82.0
80.7
80.3
78.8
81.2
70.1
37.0
18.3
30.7
Irish
0.3
0.1
0.2
0.2
0.3
0.4
0.1
0.2
0.3
0.8
0.7
0.8
Traveller Of Irish Heritage
0.1
0.0
0.0
0.1
0.1
0.1
0.1
0.1
0.0
0.1
0.1
0.1
Gypsy/Roma
0.2
0.3
0.2
0.1
0.4
0.2
0.5
0.2
0.2
0.2
0.1
0.2
White other groups
4.3
1.3
3.1
2.1
4.3
4.8
2.6
3.5
2.4
10.4
12.3
11.0
Mixed White & Black Caribbean
1.4
0.2
0.9
0.8
1.1
1.3
1.1
1.5
2.1
2.3
3.3
2.6
Mixed White & Black African
0.6
0.2
0.4
0.4
0.6
0.6
0.3
0.4
0.3
1.2
1.4
1.2
Mixed White & Asian
1.0
0.5
0.8
0.8
1.3
1.0
0.9
0.9
1.1
1.7
1.1
1.5
Mixed Other heritages
1.6
0.6
1.0
1.2
1.6
1.7
1.0
1.2
1.3
3.4
4.4
3.7
Indian
2.7
0.6
0.8
1.8
2.0
1.4
1.4
3.9
4.5
7.3
2.5
5.6
Pakistani
4.0
1.4
0.4
5.0
2.2
2.1
8.4
1.9
8.2
5.0
3.1
4.4
Bangladeshi
1.6
0.9
0.3
1.3
0.6
1.1
0.8
0.6
1.8
2.0
12.3
5.5
Any other Asian
1.6
0.7
0.6
0.9
1.7
1.1
0.9
1.1
1.2
5.8
2.6
4.7
Black Caribbean
1.3
0.0
0.3
0.3
0.3
0.6
0.4
0.6
1.8
4.3
8.9
5.9
Black African
3.3
0.5
0.8
1.3
1.6
1.9
1.3
1.5
2.3
11.3
18.0
13.6
Black Other groups
0.6
0.1
0.2
0.4
0.3
0.4
0.3
0.4
0.5
1.8
3.4
2.3
Chinese
0.4
0.3
0.3
0.4
0.4
0.4
0.3
0.3
0.3
0.7
0.8
0.7
Any Other group
1.5
0.6
0.5
1.0
0.7
0.7
0.9
0.6
1.3
4.8
7.0
5.5
Unclassified
0.9
0.4
0.9
0.7
1.0
1.1
0.6
0.8
0.8
1.5
1.4
1.4
26.7
16.7
8.4
4.0
10.6
4.9
18.0
9.6
19.3
10.1
19.7
10.5
21.2
11.8
18.8
11.3
29.9
19.9
63.0
45.5
81.7
73.6
69.3
55.0
Ethnic Minority 2013
Ethnic Minority 2003
Note: data sourced from DFE SFR 21/2013 and SFR 09/2003. Pupils of compulsory school age (516) in maintained primary and secondary schools (author’s calculation).
Across England 26.7% of pupils in England were from ethnic minority groups in 2013.
However, this varied widely, from lows of around 8%-11% in the North East and the
South West, to slightly above the national average in West Midlands (29.9%) but with
the largest concentrations in London, where minority ethnic pupils accounted for nearly
two-thirds (63.0%) of pupils in Outer London and four-fifths (81.7%) of pupils in Inner
London.
Ethnic minority population by Local Authority (LA)
Regions are high levels of aggregation and the above figures hide large variation
between Local Authorities (LA). The distribution of ethnic minority pupils across the
Local Authorities of England is presented graphically in Figures 1 and 2. Figure 1
presents the percentage of ethnic minority students aged 5-16 in primary and
secondary schools for each LA in England in 2003. Figure 2 presents the same data
from 2013. Notes on the construction of these figures are included in Appendix 3.
It is apparent from the figures that the LAs with the highest concentrations of ethnic
minority students are in inner London, Birmingham and surrounding areas in the West
Midlands, Manchester and Bradford areas, and also in Leicester, Luton and Bristol. LA
level data can be found in the DfE SFRs, though these give primary and secondary
figures separately and readers will have to combine the two to recreate the graphed
data.
19
Table 3 presents an excerpt from the data from January 2013, presenting the 25 LAs
with the highest and the 25 LAs with the lowest proportion of ethnic minority students.
Of the 25 LAs with the highest proportion of ethnic minority students, all but three are
London boroughs (the exceptions being Slough, Luton and Leicester). Full LA level
data are contained in DfE SFR 21/2013 2.
2
https://www.gov.uk/government/publications/schools-pupils-and-their-characteristics-january-2013
20
Figure 1: Percentage ethnic minority students by Local Authority: England 2003
21
Figure 2: Percentage ethnic minority students by Local Authority: England 2013
22
Table 3: Number and percentage of ethnic minority pupils: Highest and lowest LAs: 2013.
Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
LA
Redcar and Cleveland
Halton
Durham
Cumbria
St. Helens
East Riding of Yorkshire
Knowsley
Northumberland
Derbyshire
Hartlepool
Cornwall
Sefton
Rutland
Wigan
Devon
North East Lincolnshire
Isle of Wight
Barnsley
Sunderland
North Yorkshire
Shropshire
North Tyneside
Dorset
South Tyneside
Wirral
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
Croydon
Leicester
Barnet
Luton
Hammersmith and Fulham
Wandsworth
Islington
Hounslow
Camden
Lewisham
Kensington and Chelsea
Enfield
Southwark
Slough
Waltham Forest
Haringey
Redbridge
Ealing
Harrow
Hackney
Lambeth
Westminster
Tower Hamlets
Brent
Newham
Ethnic minority
pupils
620
563
2,152
2,366
892
1,834
727
1,874
4,610
624
3,283
1,943
267
2,199
4,814
1,142
922
1,642
2,078
4,590
2,076
1,597
3,222
1,146
2,856
30,559
26,941
29,869
20,907
11,170
18,129
13,752
24,408
14,156
24,016
7,376
35,182
24,331
17,043
26,508
24,400
35,527
33,402
22,858
21,624
23,677
15,632
30,253
33,537
41,302
All Classified
pupils
17,605
15,780
59,470
60,540
21,830
41,370
16,200
39,905
92,725
12,185
62,015
34,885
4,790
38,530
83,410
19,440
15,240
26,815
33,350
73,005
33,015
24,300
48,925
17,270
42,360
46,155
40,060
43,965
29,880
15,510
24,880
18,775
32,465
18,625
31,595
9,685
46,160
31,055
21,565
33,050
30,330
43,705
40,560
27,385
25,540
27,655
18,150
33,855
36,470
44,515
% ethnic
minority
3.5
3.6
3.6
3.9
4.1
4.4
4.5
4.7
5.0
5.1
5.3
5.6
5.6
5.7
5.8
5.9
6.0
6.1
6.2
6.3
6.3
6.6
6.6
6.6
6.7
66.2
67.3
67.9
70.0
72.0
72.9
73.2
75.2
76.0
76.0
76.2
76.2
78.3
79.0
80.2
80.4
81.3
82.4
83.5
84.7
85.6
86.1
89.4
92.0
92.8
Notes: Data source SFR 21/2013. Isles of Scilly and City of London excluded as they each contain only
a single school. Percentage calculated from a base of all classified pupils (99.1% of all students).
23
Ethnic trends in educational attainment
1991-2006: The Youth Cohort Study
The Youth Cohort Study (YCS) provided the first source of nationally representative
data on ethnicity and educational achievement in England. The YCS collected data on a
series of nationally representative samples of approximately 15,000 young people
shortly after they finished compulsory education at age 16. Although the YCS started in
the mid 1980's the data collected on ethnicity was very basic, identifying just four
groups (White, Black, Asian and Other). It was not until 1991 that the Asian category
was sub-divided into Indian, Pakistani and Bangladeshi, so we start our time series in
1991. The series end with the YCS Cohort 13 (Sweep 1) in 2006, which has been
combined with the Longitudinal Survey of Young People in England (LSYPE) Wave 4
sample to substantially increase sample size and reliability (DfE, 2008).
Figure 3: Percentage of 16 year olds achieving 5 or more GCSEs at grades A*-C: YCS 1991-2006.
Ethnic origin
White
Black
Asian
Indian
Pakistani
Bangladeshi
Other Asian
Other ethnic group
All Pupils
Weighted sample
1991
37
23
33
38
26
14
46
*
37
24,922
1993
43
21
36
45
24
20
50
37
42
18,020
1995
45
23
38
48
23
25
61
46
44
15,899
1997
47
29
45
54
29
33
61
47
46
14,662
1999
50
38
48
60
29
29
72
42
49
12,899
2001
52
36
52
60
40
43
64
53
51
15,714
2003
55
34
55
72
38
44
65
59
54
13,178
2006
58
50
64
72
52
57
77
56
58
19,114
Notes: data sourced from Table 4.1.1 of DfE (2008). Statistical Bulletin. YCS & LSYPE: The activities and
experiences of 16 year olds: England 2007. Results for White students indicated by the smoothed black line.
24
The results indicate that all ethnic minority groups made significant progress in the
period 1991-2006.
•
Indian pupils moved from an equivalent position to White students in 1991 to be
significantly ahead in 2006, and Other Asian pupils further increased their lead.
•
Bangladeshi pupils made the greatest progress. In 1991 just 14% of
Bangladeshi pupils achieved the benchmark of 5+ GCSE A*-C grades (5AC)
compared with 37% of White pupils. In 2006 57% of Bangladeshi pupils achieved
this benchmark against 58% of White pupils, i.e. the gap between Bangladeshi
and White pupils effectively disappeared.
•
Pakistani and Black pupils have also narrowed the gap with White pupils
substantially. Both groups doubled the proportion achieved 5AC, from 26% to
52% for Pakistani pupils and from 23% to 50% for Black pupils. Most of the
improvement, particularly for Black pupils, occurred between 2003 and 2006.
2003-2013: The National Pupil Database: 5+ GCSE A*-C
grades or equivalent (5AC)
In January 2002 the Annual School Census (ASC) was introduced which for the first
time collected individual pupil level data on the ethnicity of all children in England
attending state maintained schools. Compared to the YCS a much more differentiated
ethnic coding system was used which aligned with the ethnic categories used in the
national population census in
2001. Differentiation was made not only within the ‘Asian’ group but also within the
‘Black’ ethnic groups revealing contrasting patterns of attainment between Black
Caribbean, Black African and Black Other groups. A much more detailed picture of
achievement and change over time can be achieved through analysis of the National
Pupil Database (NPD).
2003 is a natural base year to consider given the YCS data indicate that for some
groups, particularly Black and Pakistani students, the biggest changes occurred after
2003 rather than before it. This is also an appropriate base year because a revised
ethnic coding system breaking down the ‘White’ group and introducing mixed heritage
categories was introduced in 2003, making direct comparisons with the 2002 data
problematic. Given that individual pupil level data were collected for the entire national
cohort, this also now allows us to breakdown the ethnic group results by other pupil
characteristics such as gender.
25
Figures 4 & 5 present a breakdown of the percentage of students achieving 5AC by
ethnic group and gender for the 11 years 2003 – 2013 (the data is tabulated in
Appendix 4).
26
Figure 4: Percentage 5+ GCSE A*-C or equivalent by ethnic group and gender: 2003-2013
Notes: For tabulated data see Appendix 4. For data sources see Appendix 1.
27
Figure 5: Percentage of pupils achieving 5+ A*-C GCSE or equivalent by ethnic group: 2003-2013
Notes: For tabulated data see Appendix 4. For data sources see Appendix 1.
The extent of improvement in the attainment of all ethnic groups over the period is
substantial. Half (50.7%) of all students achieved 5AC in 2003 which rose to 82.9% in
2013. There is debate about what lies behind this increase, whether it reflects a ‘real’
rise in standards or grade inflation. However, this is not essential to our focus which is
on the extent to which different ethnic groups improved, and particularly whether ethnic
groups that were underachieving relative to White British students have closed the gap.
In relation to ethnicity the results indicate:
•
The aggregation into a single ‘Black’ category in the YCS did obscure significant
differences between Black groups. Black Caribbean pupils, and in particular
Black Caribbean boys, had much lower achievement that Black African students.
Mixed White & Black Caribbean students initially (2003-2006) had somewhat
higher achievement than Black Caribbean students, but since the mid 2000’s
have an achievement profile very similar to Black Caribbean students.
•
In 2003 the odds for Black Caribbean pupils achieving 5AC were less than half
the odds for White British pupils, and for Mixed White & Black Caribbean and
Black African pupils only two-thirds the odds for White British pupils. Black
African pupils closed the gap relative to White British in 2009 and in 2013
achieve slightly above the White British average (OR=1.11). Black Caribbean
28
and Mixed White & Black Caribbean students have also improved substantially
closing the gap with White British from over 18 percentage points to just two
percentage points in 2013, although a small gap remains (OR= 0.86 for both
minority groups).
•
The YCS data indicated a big improvement for Bangladeshi students had
occurred in the years from 1991 to 2003, but in 2003 they still started out below
the average for White British students (45.5% vs. 51.3%; OR=0.79). Bangladeshi
girls achieved parity with White British girls in 2006, and Bangladeshi Boys with
White British boys in 2009. By 2013 Bangladeshi students are achieving above
the White British average (85.0% vs. 82.7%, OR=1.19). This is despite
Bangladeshi students being one of the most socio-economically deprived
communities in England (See Strand et al, 2010 for a detailed analysis of
achievement of Bangladeshi students in England).
•
Pakistani pupils have also improved substantially. In 2003 their odds of
achieving 5AC were only two thirds the odds for White British students (41.5%
vs. 51.3%, OR=0.67). However, they closed the gap steadily and achieved parity
with White British students in 2011 and currently their odds of achieving 5AC do
not significantly differ from White British students (OR=1.07).
•
Indian and Chinese pupils have consistently outperformed the White British
average. For both groups the odds of achieving 5AC are currently around twice
as high as for White British pupils (OR=1.82 and OR=2.62, respectively).
2004-2013: 5+ GCSE A*-C or equivalent including English &
mathematics (5EM)
Introduction
The above results indicate a significant and substantial closing of ethnic achievement
gaps in the last twenty years. However, there are some limitations to the 5AC indicator
analysed above. At one level the rapid rise in results with 83% of all students achieving
5AC in 2013, and averaging between 80%-90% across ethnic groups, creates a ceiling
effect. 5AC is perhaps more an indicator of a minimum standard of achievement today
than it was eleven years ago. However wider concerns were expressed by government
in the mid 2000's that the indicator was problematic since it did not require the inclusion
of passes in the core subjects of English and mathematics. As a result of this a new
threshold measure was introduced in 2006, five or more GCSE A*-C grades or
equivalent including English and mathematics (5EM). This has become the central
focus in measuring achievement and we focus on the 5EM indicator below.
29
The 5EM indicator was first introduced into performance tables in 2006. For
comparative purposes the DfE published the 5EM indicator for 2005 at the same time,
but did not publish the 5EM indicator for any student groupings prior to 2005, other than
as a single national average (SFR 41/2014). Figures for 2004 were not available as part
of the published DfE time series. However, we felt it was important to establish a full 10year time series for the analysis so we calculated our own figures direct from the
2003/04 NPD files for ethnicity, gender and FSM. Our figures agree closely with another
source we have found 3 and we believe are reliable enough for inclusion in the time
series.
Contrasting 5AC and 5EM
Before discussing the trends for 5EM we first contrast the pattern of ethnic group
differences on 5AC and 5EM for the 2004 cohort, the first year for which we have data
on both measures. Overall the proportion of students achieving 5AC in 2004 was 51.9%
while only 40.9% achieved 5EM, a difference of 11 percentage points. However the
change was not consistent across all ethnic groups. Table 4 presents the ethnic
minority Odds Ratios from 2004 for both 5AC and 5EM. For almost all ethnic minority
groups the achievement gaps are larger on 5EM than 5AC. Changes in the ORs are
greatest for Pakistani and Bangladeshi (and Indian and Chinese) students, groups
where the proportion of students with English as an additional language (EAL) is
highest, perhaps indicating a particular issue with achievement in GCSE English.
However, there were increases in the achievement gaps for Black Caribbean and Mixed
White and Black Caribbean students also.
Table 4: Ethnic group Odds Ratios for the 2004 cohort for both 5EM and 5AC measures.
Odds ratios
White British
White Other groups
Mixed White & Black Caribbean
Indian
Pakistani
Bangladeshi
Black African
Black Caribbean
Chinese
5AC
1.11
0.60
1.82
0.75
0.86
0.70
0.51
2.62
5EM
1.09
0.53
1.68
0.64
0.68
0.64
0.42
2.48
Change
-0.02
-0.07
-0.14
-0.11
-0.18
-0.06
-0.09
-0.14
The above results show how achievement gaps are sensitive to the particular measure
that defines success. Given that success in educational terms is now clearly defined by
achieving 5EM the rest of this analysis will focus on 5EM.
3
An ethnic breakdown for 5EM in 2004 was found in one DfE source (DfES, 2006, figure 28, p58) but this
did not allow any breakdowns by gender or FSM.
30
Ethnic achievement gaps on 5EM 2004-2013
Trends over the 10 years 2004-2013 are presented in Table 5 and Figure 6.
Table 5: Percentage of all students achieving 5EM by ethnic group: 2004-2013
Ethnic group
White
White British
Irish
Traveller of Irish Heritage
Gypsy / Roma
Any other White background
Mixed
White and Black Caribbean
White and Black African
White and Asian
Any other mixed background
Asian
Indian
Pakistani
Bangladeshi
Any other Asian background
Black
Black Caribbean
Black African
Any other Black background
Chinese
Any other ethnic group
All pupils
2004
41.6
46.7
27.1
8.9
43.7
27.5
36.9
57.0
41.5
54.5
31.3
32.5
48.2
23.2
31.3
24.6
63.9
37.3
40.9
2005
43.0
42.9
50.7
20.7
9.1
46.2
41.9
30.5
40.8
56.6
46.1
44.0
57.4
32.5
34.5
50.8
30.7
27.1
35.0
27.5
68.8
40.3
42.5
2006
44.4
44.3
50.1
11.1
3.9
46.8
42.8
32.6
43.1
59.4
45.2
46.1
59.1
34.6
39.0
51.6
33.6
29.5
37.5
31.2
65.8
41.7
44.0
2007
46.1
46.1
52.6
8.6
7.0
46.3
44.5
34.1
42.5
58.9
48.5
48.2
62.0
37.3
41.4
50.8
37.1
33.2
40.8
33.5
70.7
42.5
45.8
2008
48.4
48.4
57.0
7.3
6.8
46.0
47.4
38.3
46.9
58.8
51.1
50.9
65.1
40.0
45.0
52.4
40.7
36.4
43.9
39.6
69.9
44.6
48.2
2009
50.7
50.9
58.0
9.2
9.1
47.7
51.3
42.3
51.0
62.3
54.9
53.1
67.0
42.9
48.3
54.3
44.5
39.4
48.4
41.2
71.6
47.4
50.7
2010
54.8
55.0
63.4
21.8
8.3
50.6
54.6
45.3
55.6
65.2
57.8
58.0
71.3
49.1
53.7
57.6
48.9
43.5
52.8
45.8
75.1
51.2
54.8
2011
58.0
58.2
65.9
17.5
10.8
54.3
58.5
49.1
57.6
68.1
62.3
61.8
74.4
52.6
59.7
62.2
54.3
48.6
57.9
52.6
78.5
54.0
58.2
2012
58.6
58.9
66.9
16.7
9.3
52.8
59.8
52.5
59.6
67.9
62.3
62.7
74.4
54.4
62.2
61.6
54.6
49.8
58.0
50.0
76.4
56.0
58.8
2013
60.2
60.5
68.8
17.5
13.8
55.4
62.5
54.9
63.3
69.7
65.5
64.2
75.7
55.5
64.0
64.3
58.1
53.3
61.2
54.6
78.1
59.2
60.6
Odds Ratios
White Other groups
Mixed White & Black Caribbean
Indian
Pakistani
Bangladeshi
Black African
Black Caribbean
Chinese
2004
1.09
0.53
1.68
0.64
0.68
0.64
0.42
2.48
2005
1.14
0.58
1.79
0.64
0.70
0.72
0.49
2.94
2006
1.11
0.61
1.82
0.67
0.80
0.75
0.53
2.42
2007
1.01
0.61
1.91
0.70
0.83
0.81
0.58
2.82
2008
0.91
0.66
1.99
0.71
0.87
0.83
0.61
2.48
2009
0.88
0.71
1.96
0.72
0.90
0.90
0.63
2.43
2010
0.84
0.68
2.03
0.79
0.95
0.92
0.63
2.47
2011
0.85
0.69
2.09
0.80
1.06
0.99
0.68
2.62
2012
0.78
0.77
2.03
0.83
1.15
0.96
0.69
2.26
2013
0.81
0.79
2.03
0.81
1.16
1.03
0.75
2.33
31
Figure 6: Percentage of all students achieving 5EM by ethnic group: 2004-2013
The results indicate the following:
•
At the high end Indian students have further increased their advantage over
White British students and Chinese students have maintained their advantage.
The odds of students from these ethnic groups of achieving 5EM in 2013 are
over twice the odds for White British students.
•
Bangladeshi students have substantially increased in achievement from well
below the White British average in 2004 (OR=0.68) to above the White British
average in 2013 (OR=1.16).
•
Black African students have also show substantial increases in their
educational achievement and there is no longer any gap relative to White British,
the OR changing from OR=0.65 in 2004 to OR=1.03 in 2013.
•
There has been a significant closing of the gap for Black Caribbean, Mixed
White & Black Caribbean and Pakistani students. For example the odds for
Black Caribbean students achieving the threshold were less half the odds for
White British in 2004 (OR=0.43) but in 2013 this had narrowed to OR=0.75. For
Mixed White & Black Caribbean students there has been a change from
OR=0.53 to OR=0.79 and for Pakistani students from OR=0.64 to OR=0.81.
However, the odds of achieving 5EM for all three groups are still around 20%
lower than the average for White British students.
32
•
Only one ethnic group has seen performance decline relative to White British
over the period. White other groups have moved from above the White British
average in 2004-2006 to below the White British average from 2008 onwards
and in 2013 the OR=0.81 is similar to that for Black Caribbean, Mixed White &
Black Caribbean and Pakistani students. It should be remembered that we are
looking at different cohorts of pupils in the time series and we saw in the
demographics presented earlier that there has been a considerable increase in
the White Other groups from 135,000 (2.1% of the population) in 2003 to
287,000 (4.3% of the population) in 2013.
Capped (Best 8) points scores
Any threshold indicator, be it 5AC or 5EM, can be sensitive to changes in results
around the threshold. For this reason a points score averaged across all grades and
across a range of subjects can have advantages. The capped (Best 8) points score has
been used by the DfE for a number of years as an outcome indicator in value added
calculations and in school performance tables. The indicator converts grades in GCSEs
and a range of other qualifications to points scores and sums the scores for the highest
eight GCSE, or equivalent examinations, achieved by the student 4
The Best 8 Points score has been published in school performance tables and used in
school value added calculations but has not formed a routine part of DfE reports on
GCSE achievement by pupil characteristics, there are therefore no routinely published
statistics for the Best 8 score in relation to ethnicity or other pupil background
characteristics. There was insufficient time to collate and analyse pupil level data from
the various NPD files to calculate figures for all years. However, NPD pupil level data
for 2004, 2007, 2009 and 2013 were available, and we calculated Best 8 scores for
these years. Table 6 and Figure 7 present the mean and standard deviation of Best 8
score for each ethnic group. In addition a measure of effect size (Cohen's D) is
presented. This expresses the difference in mean score between each ethnic group and
White British students as a proportion of the pooled (here all pupil) standard deviation.
The results follow a very similar pattern to the 5EM outcome:
4
•
Indian and Chinese pupils consistently exceed the White British average by a
wide margin (ES=0.39 and 0.60 respectively in 2013).
•
Bangladeshi pupils have improved and now have a mean score slightly higher
than the White British average (ES=0.09). Both Pakistani and Black African
students have improved substantially and their mean score in 2013 the same as
the White British average.
For details on the points awarded to different qualifications see the DfE Points Score document at
http://www.education.gov.uk/schools/performance/secondary_13/Point_Score_Document_final.pdf
33
•
Black Caribbean and Mixed White & Black Caribbean were the lowest
performing groups in 2004 (ES=-0.35 and -0.29 respectively) and continue to be
the lowest achieving groups in 2013. However, the size of these gaps has more
than halved to just ES= -0.15 and -0.12 respectively in 2013. Both gaps would be
considered 'small' according to Cohen's criteria 5. They are also small compared
to other equity gaps. For example in 2013 the gender gap (ES=0.29) was twice
as large as the ethnic gap, and the FSM gap (ES=0.61) four times larger than
the ethnic gap.
Table 6: Best 8 points score by ethnic group: Selected years 2004-2013
Mean Best 8 points score
Ethnic group
White British
White Irish
Traveller of Irish Heritage
Gypsy / Roma
White Other background
Mixed White & Black African
Mixed White & Caribbean
Mixed White & Asian
Mixed other background
Indian
Pakistani
Bangladeshi
Any other Asian background
Black African
Black Caribbean
Black Other background
Chinese
Any other ethnic group
Unclassified
All pupils
2004
284.6
294.2
194.7
155.7
287.3
269.4
253.1
318.1
283.1
321.5
273.0
280.1
301.2
262.6
247.2
245.9
335.6
268.1
269.9
283.6
2007
295.7
300.8
124.0
132.4
287.8
290.1
268.6
316.5
297.6
334.6
284.3
294.1
296.3
283.1
273.7
271.9
348.1
283.3
280.0
295.0
2009
320.0
328.4
182.1
193.6
313.1
318.3
303.0
340.4
326.6
353.9
312.8
318.8
326.0
316.5
304.5
304.3
366.9
313.9
314.9
320.0
2013 Change
341.1
56.5
351.1
56.9
205.9
11.2
209.9
54.3
338.3
51.0
344.4
75.1
330.8
77.8
358.4
40.4
347.8
64.7
373.0
51.5
341.8
68.8
348.2
68.2
354.1
52.9
342.1
79.5
329.0
81.8
333.6
87.7
390.4
54.9
345.1
77.0
338.2
68.4
342.0
58.4
Standard Deviation
2004
107.4
111.0
146.0
123.3
120.6
118.7
111.7
106.0
118.0
88.8
100.6
100.6
116.8
110.5
104.0
109.7
114.3
127.5
114.0
108.3
2007
106.7
116.0
130.4
124.8
123.7
112.0
112.8
110.5
114.3
91.2
104.6
102.7
124.9
107.9
101.2
107.3
111.0
123.6
115.5
107.9
2009 2013
92.4 82.3
95.4 84.0
128.1 135.5
115.6 128.6
104.1 92.3
96.0 80.7
95.4 83.3
94.0 80.7
94.9 82.9
78.3 69.3
89.6 78.0
89.3 76.5
98.1 83.9
89.2 75.6
86.4 76.3
93.1 79.8
97.4 74.8
104.3 85.5
97.9 91.3
93.0 82.6
Effect Size
2004
0.09
-0.83
-1.19
0.03
-0.14
-0.29
0.31
-0.01
0.34
-0.11
-0.04
0.15
-0.20
-0.35
-0.36
0.47
-0.15
-0.14
-
2007
0.05
-1.59
-1.51
-0.07
-0.05
-0.25
0.19
0.02
0.36
-0.11
-0.01
0.01
-0.12
-0.20
-0.22
0.49
-0.11
-0.15
-
2009
0.09
-1.48
-1.36
-0.07
-0.02
-0.18
0.22
0.07
0.37
-0.08
-0.01
0.07
-0.04
-0.17
-0.17
0.50
-0.07
-0.05
-
Note: Effect Size (Cohen's D) is the difference in mean score between each ethnic group and White
British divided by the all pupils standard deviation. Ethnic groups in bold are the nine core ethnic groups.
5
Cohen (1988) suggests benchmarks for Cohen's D of around 0.20, 0.50 and 0.80+ to indicate small, medium and
large effect sizes respectively. However these are rather arbitrary rules of thumb and it is good practice to
compare the size of the effect to the size for a range of other variables on the same outcome, for example in this
case to compare the size of ethnic gaps against the size of gender or FSM gaps.
34
2013
0.12
-1.64
-1.59
-0.03
0.04
-0.12
0.21
0.08
0.39
0.01
0.09
0.16
0.01
-0.15
-0.09
0.60
0.05
-0.03
-
Figure 7: Best 8 points score by ethnic group: Selected years 2004-2013
Source: Authors own analysis of NPD.
Overview across the three indicators
Overall the pattern of results over the last 10 years is broadly consistent across different
indicators:
•
Indian and Chinese students have consistently scored substantially above the
White British average
•
Bangladeshi, Black African and Pakistani students results have improvement
substantially and achievement gaps relative to White British students have been
eliminated, and in some cases reversed
•
Black Caribbean and Mixed White & Black Caribbean students have also
shown very strong improvement, from being half as likely and White British
students to achieve the benchmarks of educational success in the early 2000’s
to near parity in 2013, although stubborn gaps do remain
•
White Other students generally scored above the White British average in the
early 2000’s but in recent years have been scoring below the White British
average, with a similar gap to Black Caribbean and Mixed White and Black
Caribbean students.
35
There are some ethnic groups that have contrasting achievement gaps across the three
indicators. Pakistani pupils perhaps show the biggest differences. There is no Pakistani
achievement gap on Best 8 score or on the 5AC indicator, but some lower achievement
on the 5EM outcome (OR=0.81). A similar, though less extreme, contrast across
indicators is also seen for White Other students. The extent of underachievement for
the White Other groups is also negligible on Best8 and very small on 5AC (OR=0.87)
but a little larger on 5EM (OR=0.81).
It may be that GCSE English is a particular challenge for some students in these two
ethnic groups, and certainly a recent analysis of the 2013 NPD in relation to EAL and
ethnicity (Strand, Malmberg & Hall, 2015) identified English as the subject with the
largest EAL gap, and Pakistani and White Other students among the groups with the
largest EAL gap. This suggests a particular focus on GCSE English for EAL pupils from
these two ethnic groups, and particularly for those who are new arrivals in England, will
be important in addressing the 5EM achievement gap.
36
Trends in the relative size of ethnic, gender and FSM gaps
We have looked above at the absolute size of the achievement gap for ethnic
minority groups relative to White British students over the last ten years and seen
how these have reduced. This section evaluates the relative size of ethnic, gender
and FSM achievement gaps and whether the ethnic achievement gap has got larger
or smaller relative to gender and FSM gaps. We index the ethnic gap as the
difference between the average for White British students and the average for the
lowest achieving of the core ethnic groups in 5EM, which in all years was Black
Caribbean students. Gender and FSM gaps are straightforward contrasts between
girls and boys and between pupils not entitled and those entitled to FSM. Table 7
and Figure 8 present the data.
Table 7: Trends in ethnic, gender and FSM achievement gaps at age 16 over time: 2004-2013
Student grouping
Ethnicity
White British
Black Caribbean
Gap (% points)
Odds Ratio (OR)
Socio-economic
FSM
Not FSM
Gap (% points)
Odds Ratio (OR)
Gender
Boys
Girls
Gap (% points)
Odds Ratio (OR)
2004 2005 2006
2007 2008 2009 2010 2011 2012 2013
40.9
22.8
18.1
2.3
42.9
27.1
15.8
2.0
44.2
29.2
15.0
1.9
45.8
32.7
13.1
1.7
48.0
35.9
12.1
1.6
50.9
39.4
11.5
1.6
55.0
43.5
11.5
1.6
58.2
48.6
9.6
1.5
58.9
49.8
9.1
1.4
60.5
53.3
7.2
1.3
16.8
44.8
28.0
4.0
18.0
46.4
28.4
3.9
19.6
47.7
28.1
3.7
21.4
49.3
27.9
3.6
23.8
51.7
27.9
3.4
26.6
54.2
27.6
3.3
30.9
58.5
27.6
3.2
34.6
62.0
27.4
3.1
36.3
62.6
26.3
2.9
37.9
64.6
26.7
3.0
37.1
44.8
7.7
1.4
38.4
46.7
8.3
1.4
39.7
48.0
8.3
1.4
41.4
49.6
8.2
1.4
43.8
51.9
8.1
1.4
47.1
54.4
7.3
1.3
51.1
58.6
7.5
1.4
54.6
61.9
7.3
1.4
54.2
63.7
9.5
1.5
55.6
65.7
10.1
1.5
Note. In the above table we have expressed the OR for ethnicity as how much higher the odds of
White British students achieving 5EM are compared to Black Caribbean students, so that ethnic,
gender and FSM gaps are all expressed as advantages. However the OR=1.34 (odds for White
British relative to Black Caribbean) is identical to the OR=0.75 (odds for Black Caribbean relative to
White British) they just reverse the base or reference group.
In terms of simple trends:
•
As we saw earlier there has been a substantial shrinking of the ethnic gap.
The odds of success for White British students remain higher than for Black
Caribbean students but have narrowed sharply with the OR reducing from
2.3 to 1.3.
•
The relative advantage for girls was stable up until 2011 at an OR of around
1.4, but has widened very slightly in the last two years due to a stalling of the
37
increase in 5EM for boys compared with a continued large increase for girls
(this is seen most clearly in Figure 9).
•
The FSM gap has narrowed over the period with the OR declining from 4.0 to
3.0, but it remains by far the largest achievement gap.
38
Figure 8: Trends in ethnic, gender and FSM achievement gaps at age 16 over time: 2004-2013
Notes: In the ethnicity graph the numbers displayed show the proportion achieving 5EM for White British students and for Black Carib group).
39
Importantly there are significant changes over time in the relative size of the
achievement gaps. Looking at the percentage point gaps, in 2004 the ethnic gap was
18.1 percentage points, more than twice as large as the gender gap (7.7% points) and
half the size of the FSM gap (28% points). However, by 2013 this had changed
substantially. Now the ethnic gap (7.2% points) is actually smaller than the gender gap
(10.1% points), and about one-quarter of the size of the FSM gap (26.7% points).
We see the same pattern for Best8 points score. Figure 9 present the effect sizes in
2004 and 2013. In 2004 the ethnic gap (taking Black Caribbean students as the lowest
scoring group) was larger than the gender gap, but in 2013 the ethnic gap is half the
size of the gender gap. Equally while in 2004 the ethnic gap was half the size of the
FSM gap, in 2013 it is just one-quarter the size of the FSM gap. These are the same
proportions we see from a comparison of the gaps using the 5EM measure.
Figure 9: Cohen's D effect size for gender, ethnic and FSM Achievement gaps in 2004 and 2013
We conclude that ethnic achievement gaps have not just narrowed in absolute terms,
they have also declined substantially relative to the gender and FSM gaps.
Achievement by combinations of ethnic group, entitlement to
FSM and gender
Having explored the overall trends in relation to ethnicity and achievement, this section
explores the extent to which ethnic trends may vary across other demographic
variables, specifically entitlement to FSM or gender. For example is the improvement in
the results for Black Caribbean students similar for both boys and girls, or among
students entitled to FSM and those not entitled to FSM? Given the relative consistency
in ethnic gaps across the three examination measures we focus on the 5EM indicator,
40
since it is has been, and is currently, the main benchmark of success at age 16 and we
have data for every year in the time series through DfE SFRs.
Presenting the number of students as well as performance in a time series analyses
over 10 years would involve a large number of additional cells. To give an indication of
the size of the ethnic, FSM and gender combinations pupil numbers for 2013 are shown
in Appendix 2. When breaking results down by ethnic group, FSM and gender
simultaneously numbers in some cells can become relatively small, even for a national
cohort of over 570,000 students. For example there are only 83 Chinese girls entitled to
FSM in 2013. While there may be greater year to year variability for such small groups,
generally numbers are sufficiently high to provide robust estimates.
Pupils not entitled to FSM
Figure 10 presents the 5EM results for pupils not entitled to FSM by ethnic group and
gender (Appendix 5 tabulates the data), while Table 9 and Figure 11 combine boys and
girls together. As would be expected results for pupils not entitled to FSM are slightly
higher than the results for all pupils (see Figure 6), but the pattern of performance by
ethic group and the profile of change over time is very similar, so we will not reiterate
the previous bullet point summary. This is perhaps not surprising given that pupils not
entitled to FSM are a large majority of the cohort (85% of all pupils in the 2013 Y11
cohort).
We can look here at contrasts by ethnicity and gender. The proportion of girls achieving
5EM is around 10% points higher than for boys and broadly consistent across all ethnic
groups in 2013. The only notable contrast by ethnicity and gender over time is in the
particularly strong relative improvement of Black boys. In 2004, the achievement gaps
for Black African boys (OR=0.72) and Black Caribbean boys (OR=0.37) relative to
White British boys were much larger than the gaps for Black African girls (OR=0.85)
and Black Caribbean girls (OR=0.49) relative to White British girls. However, in 2013
both Black African boys and girls are achieving above the White British average
(OR=1.09 and OR=1.05 respectively) and Black Caribbean boys and girls have a
similar gap (OR=0.69 and OR=0.75 respectively), i.e. Black boys have made
proportionally more improvement 2004-2013 than Black girls.
The big change to previous results occurs when we look at ethnic gaps among pupils
entitled to FSM, which are radically different from the results above and to which we
now turn.
41
Figure 10: GCSE 5EM for pupils not entitled to FSM by ethnic group and gender: 2004-2013
Notes: for tabulated data and OR's see Appendix 5
42
Table 8 & Figure 11: GCSE 5EM for all pupils not entitled to FSM by ethnic group: 2004-2013
Ethnic group
White
White British
Irish
Traveller of Irish Heritage
Gypsy / Roma
Any other White background
Mixed
White and Black Caribbean
White and Black African
White and Asian
Any other mixed background
Asian
Indian
Pakistani
Bangladeshi
Any other Asian background
Black
Black Caribbean
Black African
Any other Black background
Chinese
Any other ethnic group
All pupils
2004
45.0
52.3
39.4
14.3
49.2
32.7
41.7
62.8
47.6
57.3
37.2
37.5
53.7
26.4
39.4
28.2
65.0
44.7
44.7
2005
46.6
46.5
56.8
31.9
11.6
51.3
47.7
35.1
46.8
61.5
52.0
50.6
60.3
37.9
39.9
56.1
35.7
30.0
42.7
32.8
69.9
47.0
46.4
2006
47.8
47.7
56.5
x
x
51.5
48.5
37.5
48.4
64.2
51.0
51.6
61.7
38.9
43.4
56.2
38.3
32.7
44.6
35.1
67.3
48.2
47.7
2007
49.4
49.4
57.9
13.3
8.4
50.1
49.8
39.0
47.3
64.4
53.1
53.5
64.4
41.4
46.7
54.3
41.2
35.9
47.0
37.0
71.8
47.0
49.4
2008
51.7
51.7
62.5
x
7.7
49.0
52.1
42.7
52.4
63.1
55.3
55.3
67.2
43.7
48.0
55.0
44.9
39.2
49.9
42.8
70.8
49.2
51.7
2009
54.1
54.3
64.0
x
10.4
50.3
56.3
46.8
56.7
67.0
59.7
57.5
69.2
46.9
52.8
55.9
49.0
42.1
55.1
44.3
71.7
50.0
54.2
2010 2011
58.4 61.8
58.6 62.0
69.1 71.7
27.6 25.0
9.5 11.9
52.8 56.3
59.5 63.2
50.0 53.9
60.4 61.7
69.7 72.5
62.2 66.6
62.0 65.7
73.2 76.3
53.0 56.8
56.6 62.6
60.1 64.0
53.2 58.8
46.5 52.0
58.8 63.6
50.1 57.1
75.8 78.9
54.5 56.5
58.5 62.0
2012
62.3
62.7
72.4
23.6
9.9
54.6
64.7
58.0
64.1
72.0
66.4
65.9
76.2
57.5
64.5
63.7
58.6
52.8
62.9
54.0
77.1
58.0
62.6
2013
64.2
64.5
74.2
25.0
18.0
57.3
67.5
60.7
67.7
73.8
69.8
67.4
77.2
58.8
67.0
66.4
62.5
57.0
66.2
59.6
78.2
62.7
64.6
Odd-Ratio
White Other groups
Mixed White & Black Caribbean
Indian
Pakistani
Bangladeshi
Black African
Black Caribbean
Chinese
2004
1.18
0.59
1.64
0.72
0.73
0.79
0.44
2.27
2005
1.21
0.62
1.75
0.70
0.76
0.86
0.49
2.67
2006
1.16
0.66
1.77
0.70
0.84
0.88
0.53
2.26
2007
1.03
0.65
1.85
0.72
0.90
0.91
0.57
2.61
2008
0.90
0.70
1.91
0.73
0.86
0.93
0.60
2.27
2009
0.85
0.74
1.89
0.74
0.94
1.03
0.61
2.13
2010 2011
0.79 0.79
0.71 0.72
1.93 1.97
0.80 0.81
0.92 1.03
1.01 1.07
0.61 0.66
2.21 2.29
2012
0.72
0.82
1.90
0.80
1.08
1.01
0.67
2.00
2013
0.74
0.85
1.86
0.79
1.12
1.08
0.73
1.97
43
Pupils entitled to FSM
Figure 12 presents the results for pupils entitled to FSM broken down by ethnic group
and gender (the data is tabulated in Appendix 6). Table 9 and Figure 13 present the
combined data for boys and girls.
•
The most striking feature, in contrast to the results for pupils not entitled to
FSM, is that among those entitled to FSM all ethnic minority groups
achieve greater success than White British pupils. In 2013 relative to White
British students the odds for Chinese pupils achieving 5EM were 6.9 times
greater, for Indian students 3.4 times greater, for Bangladeshi 3.0 times, for
White other 1.6 times, for Black Caribbean 1.5 times and even for the lowest
achieving minority group, Mixed White & Black Caribbean students, the odds of
achieving 5EM were 1.26 times (26%) higher than for White British students.
•
Many ethnic minority groups have pulled even further ahead of White
British students between 2004 and 2013. Between 2004 and 2013 the OR for
Mixed White & Black Caribbean students increased from OR=0.97 to OR=1.26,
for Black Caribbean students it increased from OR=0.98 to OR=1.53, for Black
African students from OR=1.44 to OR=2.22 and for Bangladeshi students from
OR=2.52 to OR=3.0. So the gap between White British FSM students and ethnic
minorities FSM students is large and increasing: White British FSM students are
not only the lowest achieving they seem to be increasingly so over time.
•
There is very little difference between the ethnic trends for boys and girls.
The only notable point is that while Black Caribbean and Mixed White & Black
Caribbean girls started pulling ahead of White British girls in 2005, this trend did
not emerge for Black Caribbean and Mixed White & Black Caribbean boys until
2008. Overall, the trends are highly similar for both boys and girls.
•
Within the ‘Black’ group there is increasing divergence between Black
African students on the one hand and Black Caribbean and Mixed White &
Black Caribbean students on the other, among both boys and girls. The gap
between Black African and Black Caribbean students has increased from 5.2%
points in 2004 to 9.2% points in 2013 and for Mixed White & Black Caribbean
students the percentage point gap has increased from 5.4% in 2004 to 13.9%
percentage points in 2013.
44
Figure 12: GCSE 5EM for pupils entitled to FSM by ethnic group and gender: 2004-2013
Notes: for tabulated data and OR's see Appendix 6.
45
Table 9: 5+ A*-C GCSE (Inc. English & Maths) for all pupils entitled to FSM by
ethnic group: 2004-2013
Ethnic group
White
White British
Irish
Traveller of Irish Heritage
Gypsy / Roma
Any other White
background
Mixed
White and Black
Caribbean
White and Black African
White and Asian
Any other mixed
background
Asian
Indian
Pakistani
Bangladeshi
Any other Asian
background
Black
Black Caribbean
Black African
Any other Black
background
Chinese
Any other ethnic group
All pupils
2004
14.1
19.5
2.2
1.9
2005
15.1
14.7
21.4
2.4
6.1
2006
16.2
16.0
19.5
x
x
2007
17.7
17.4
24.1
x
x
2008
19.5
19.1
24.6
x
5.4
2009
21.9
21.5
24.0
x
6.9
2010
25.8
25.3
29.1
16.7
6.2
2011
29.3
28.8
33.6
12.3
8.3
2012
30.9
30.5
32.7
11.7
8.5
2013
32.7
32.3
38.5
12.9
9.2
20.1
-
23.7
21.2
23.6
22.0
26.7
24.6
29.4
27.9
31.9
30.7
37.0
34.9
41.6
39.5
41.9
41.3
43.8
43.9
13.7
24.3
27.6
17.8
23.7
27.7
18.0
25.6
31.4
19.4
27.6
29.6
23.1
28.1
34.3
27.4
29.1
35.9
30.0
37.8
38.6
33.7
43.6
42.5
36.0
44.7
46.3
37.5
48.6
47.9
18.5
35.3
22.5
29.3
22.4
28.6
37.4
24.1
30.5
22.6
32.0
39.5
27.1
35.3
28.7
33.7
41.9
29.5
36.4
31.6
37.6
45.9
32.3
41.7
33.4
39.6
48.0
34.2
43.0
38.9
45.7
55.0
40.6
50.3
43.9
49.5
57.0
42.9
56.2
44.5
51.8
57.9
46.5
58.6
48.0
52.8
61.5
46.8
59.2
29.8
13.9
19.1
32.7
20.3
18.8
22.4
34.4
23.1
19.5
25.5
36.0
27.1
24.2
29.2
40.5
30.4
26.2
32.1
45.9
33.8
29.5
35.6
45.3
39.1
33.1
42.1
51.8
44.3
37.8
47.2
49.6
45.6
40.2
48.4
52.4
48.2
42.2
51.4
16.3
55.4
24.2
16.8
14.8
58.6
27.6
18.0
20.7
53.8
29.7
19.6
23.5
60.7
33.4
21.4
30.7
63.1
35.5
23.8
32.7
70.8
42.2
26.6
34.9
68.4
44.2
30.9
41.3
73.5
48.5
34.6
40.0
68.2
51.2
36.3
43.1
76.8
51.5
37.9
Odd-Ratio
White Other groups
Mixed White & Black
Caribbean
Indian
Pakistani
Bangladeshi
Black African
Black Caribbean
Chinese
2004
1.53
2005
1.80
2006
1.62
2007
1.73
2008
1.76
2009
1.71
2010
1.73
2011
1.76
2012
1.64
2013
1.63
0.97
3.32
1.77
2.52
1.44
0.98
7.57
1.26
3.47
1.84
2.55
1.68
1.34
8.21
1.15
3.43
1.95
2.86
1.80
1.27
6.11
1.14
3.42
1.99
2.72
1.96
1.52
7.33
1.27
3.59
2.02
3.03
2.00
1.50
7.24
1.38
3.37
1.90
2.75
2.02
1.53
8.85
1.27
3.61
2.02
2.99
2.15
1.46
6.39
1.26
3.28
1.86
3.17
2.21
1.50
6.86
1.28
3.13
1.98
3.23
2.14
1.53
4.89
1.26
3.35
1.84
3.04
2.22
1.53
6.94
Notes: 2004 data authors own calculations from NPD. 2005-2013 taken from the relevant DfE
SFRs.
46
Figure 13: 5+ A*-C GCSE (Inc. English & Maths) for all pupils entitled to FSM by
ethnic group: 2004-2013
47
Check against Best 8 score
For completeness Table 10 and Figure 14 present a breakdown of Best 8 score by
ethnic group and entitlement to FSM to see whether the trends discussed above for
5EM are also reflected in Best 8 score.
Table 10: Best 8 points score by ethnic group and entitlement to FSM
Ethnic group
White British
White Other groups
Mixed White & Carib.
Indian
Pakistani
Bangladeshi
Black African
Black Caribbean
Chinese
2004
200.8
229.0
212.4
283.5
255.6
276.1
237.4
222.2
319.5
Effect size
White Other groups
Mixed White & Carib.
Indian
Pakistani
Bangladeshi
Black African
Black Caribbean
Chinese
0.24
0.10
0.71
0.47
0.65
0.32
0.18
1.02
Entitled to FSM
2007
2009
215.9
253.3
252.5
285.5
228.7
268.2
295.1
323.8
267.6
297.1
287.2
310.7
258.8
294.2
251.0
287.5
324.8
360.4
0.31
0.11
0.67
0.44
0.60
0.36
0.30
0.92
0.31
0.14
0.67
0.42
0.55
0.39
0.33
1.02
2013
287.7
310.0
300.9
344.6
328.7
337.5
326.3
311.3
372.0
Change
0.22
0.13
0.56
0.40
0.49
0.38
0.23
0.83
-0.02
0.03
-0.15
-0.07
-0.16
0.07
0.05
-0.19
86.9
81.0
88.5
61.2
73.1
61.3
88.9
89.1
52.6
Not Entitled FSM
2004
2007
295.0
305.0
300.9
294.5
268.6
281.8
327.0
339.3
284.6
293.0
286.4
301.5
279.3
295.8
256.0
280.5
337.8
350.6
0.06
-0.26
0.31
-0.10
-0.08
-0.15
-0.38
0.42
-0.10
-0.23
0.33
-0.12
-0.03
-0.09
-0.24
0.44
2009
327.5
317.5
313.4
357.3
319.8
325.6
328.2
309.1
367.5
2013
348.7
343.1
340.8
376.0
346.9
355.0
350.1
335.0
391.9
Change
-0.11
-0.16
0.34
-0.09
-0.02
0.01
-0.21
0.45
-0.07
-0.10
0.36
-0.02
0.08
0.02
-0.18
0.57
-0.13
0.15
0.05
0.08
0.17
0.17
0.20
0.15
53.7
42.2
72.2
49.0
62.3
68.6
70.8
79.0
54.1
The pattern of closing ethnic achievement gaps over time among non-FSM pupils is
equally apparent for Best 8 score as for 5EM. The pattern of much higher achievement
by ethnic minority FSM pupils compared to White British FSM pupils is also very clear.
However, an increase in the advantage of ethnic minority FSM pupils between 2004 and
2013 is not evident. Since the Best 8 points score advantage for FSM ethnic minority
students has not increased, the increased advantage in the proportion achieving 5EM
presumably reflects a greater likelihood of achieving A*-C grades in GCSE English and
mathematics in particular.
48
Figure 14: Best 8 points score by ethnic group and entitlement to FSM
49
Adjusting ethnic achievement gaps for socio-economic
factors
It should be noted that the above data are unadjusted, they take no account of the fact
that levels of socio-economic disadvantage are considerably higher among ethnic
minority groups compared to the White British group. For example Figure 15 shows the
proportion of each ethnic group entitled to FSM and Figure 16 shows the proportion in
each Income Deprivation Affecting Children Index (IDACI) score quintile (from very low
to very high deprivation). On both measures the considerably greater poverty within
ethnic minority groups is clear.
Studies that use regressions methods to adjust ethnic results to account for socioeconomic differences suggest that almost all ethnic minority groups achieve better than
the White British students of comparable SES at age 16. For example Strand (2014a)
uses the Longitudinal Study of Young People in England (LSYPE) to explore ethnic
gaps in achievement at GCSE in 2006 after adjusting for a socio-economic status (SES)
factor created from a range of socio-economic variables (parental social class, parental
educational qualifications, neighbourhood deprivation, entitlement to FSM and home
ownership). All ethnic minority groups achieve at least as well and frequently
substantially better than the White British students, with the single exception of middle
and high SES Black Caribbean boys (see Figure 17).
50
Figure 15: Percentage of students age 5-16 entitled to a FSM: England 2011
Figure 16: Percentage of students in each of five score bands based on IDACI: England 2011
Source: Both graphs authors own calculations based on all students aged 5-16 in January 2011 (See
Strand, 2012).
51
Figure 17: GCSE results at age 16 adjusted for Socio-economic Status (SES)
Notes: (1). Based on the LSYPE nationally representative sample of 15,000 students age 16 in 2006. (2) The outcome (total points score) is a measure of
achievement based on all examinations completed by the young person at age 16, and is expressed on a scale where 0 is the mean (average) score for all
Young People at age 16 and two-thirds of students score between -1 and 1. (3). The SES measure also has a mean (average) of zero and the effects for low
SES are estimated at -1SD and of high SES at +1SD. Source: Strand (2014a) Figure 4 for full details.
52
Ethnic achievement gaps at primary school
It should not be assumed that achievement gaps observed at age 16 are necessarily
the result of influences occurring during the secondary school phase. The gaps may
reflect earlier gaps from primary school or the early years. To better understand the
origins of the low achievement of White British students entitled to FSM at age 16 it is
important to evaluate data from earlier in the educational system.
Figure 18 displays the age 16 5EM data is a form that (i) highlights the achievement of
each ethnic and FSM combination and (ii) facilitates consistent comparison with the
results from primary school assessments which follow. As we have seen earlier, at age
16 White British students on FSM are the lowest achieving group, and the FSM gap for
White British students (indicated by the red bars) is the largest of all ethnic groups.
Figure 18: Age 16 5EM results by ethnic group and FSM: England 2013
Source: Authors graphic based on data from DfE SFR 05/2014. Irish Traveller and Gypsy
Roma Travellers not included because of small sample size. Black horizontal line indicates the
England average.
53
End of Key Stage 2 (age 11)
Figure 19 presents the national end of key stage 2 (KS2) test results at age 11. The
figure plots the KS2 threshold measure, which is the percentage of pupils gaining level
4 or above in all three domains of reading, writing and mathematics. Again there is a
separate bar plotted for each ethnic group and FSM combination.
White British FSM pupils (58%), along with White Other FSM pupils (56%), are the
lowest achieving groups 6. It is apparent that the low achievement of White British FSM
is not particular to the secondary phase and a substantial gap exists at age 11,
although there is evidence that the gap grows even further during the secondary phase
(e.g. Strand, 2014a).
Figure 19: Age 11 results by ethnic group and FSM: England 2013
Source: Authors graphic based on tabulated data presented in Appendix 7. Irish Traveller and Gypsy
Roma Travellers not included because of small sample size. Data source = DfE SFR 51/2013. Horizontal
line indicates the England average.
6
We know that for many pupils in the White Other group their low achievement at age 11 reflects EAL issues
associated with recent arrival in England and low fluency in the English language. Indeed when these factors are
taken into account they make above average progress age 11-16 (See Strand et al. 2015 for a full analysis).
54
End of Foundation Stage (age 5)
The earliest national collection of data on educational attainment is the Early Years
Foundation Stage Profile (EYFSP) completed at the end of reception year when
children are aged 5. Figure 20 presents the proportion of pupils rated as achieving a
good level of development (GLD) 7. Again there is a separate bar plotted for each ethnic
group and FSM combination.
Figure 20: Age 5 results by ethnic group and FSM: England 2013
% with a 'good level of development'
65
60
55
50
45
40
35
30
25
20
Source: See Appendix 8 for tabulated data. Authors graphic based on data from DFE SFR 47/2013. Horizontal line
indicates the England average.
White British pupils entitled to FSM (35%), along with White Other FSM (34%) and
Pakistani FSM (34%) pupils are the lowest achieving groups. However, the results for
Pakistani pupils, as for White Other groups, are related to the high number of pupils
with EAL within these two groups. A recent analysis of the NPD (Strand, Malmberg &
Hall, 2015) shows that EAL gaps are largest in the EYFSP at age 5 (GLD OR=0.67)
and more or less disappear by age 16 (5EM OR=0.90).
7
The GLD measure is the most widely used single measure of child development in the early years.
Children have been defined as reaching a GLD and the end of the EYFS if they achieved at least
the expected level in the early learning goals in the prime areas of learning (personal, social &
emotional development; physical development; and communication and language) and in the
specific areas of language and literacy. See DfE SFR 47/2013 p13.
55
The above results indicate that any analysis of the drivers of the low educational
achievement of White British pupils on FSM at age 16 must consider factors operating
in the first five years of the child's life.
Other educational outcomes
This report has focussed on ethnic group differences in educational achievement. Data
on two other educational outcomes, identification of special educational needs (SEN)
and exclusion from school, are presented in Appendix 9.
56
References
Burgess, S. (2014). Understanding the success of London's schools. University of
Bristol: Centre for Market and Public Organisation (CMPO).
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). New
Jersey: Lawrence Erlbaum.
DCSF (2007). Ensuring the attainment of Black pupils (00750-2007-FLR-EN). London:
Department for Children Schools and Families.
DfE (2008). Youth Cohort Study and Longitudinal Study of Young People in England:
The activities and experiences of 16 year olds: England 2007. London: Department for
Education.
DfE (2009). Youth Cohort Study and Longitudinal Study of Young People in England:
The activities and experiences of 17 year olds: England 2008. London: Department for
Education.
DfE (2013). Schools, pupils and their characteristics January 2013 (SFR 21/2013).
London: Department for Education. Downloaded from WWW at
https://www.gov.uk/government/publications/schools-pupils-and-their-characteristicsjanuary-2013 [accessed 04/07/13].
DfE (2013). Early years foundation stage profile attainment by pupil characteristics,
England 2013 (SFR 47/2013). London: Department for Education.
DfE (2013). National Curriculum assessments at Key Stage 2 in England, 2013
(Revised) (SFR 51/2013). London: Department for Education.
DfE (2013). GCSE and equivalent attainment by pupil characteristics in England
2011/12 (SFR 04/2013). London: Department for Education.
DfE (2014). GCSE and equivalent attainment by pupil characteristics in England
2012/13 (SFR 05/2014). Retrieved from
https://www.gov.uk/government/publications/gcse-and-equivalent-attainment-by-pupilcharacteristics-2012-to-2013.
DfEE (2000). Removing the barriers: Raising achievement levels for minority ethnic
pupils: Key points for schools. London: Department for Education and Employment
DfES (2003). Aiming High: Raising the achievement of ethnic minority pupils. London:
Department for Education and Skills.
DfES (2004). Aiming High: Supporting effective use of EMAG. (DfES-0283-2004).
London: Department for Education and Skills.
57
DfES (2006). Ethnicity and education: The evidence on minority pupils aged 5-16.
London: Department for Education and Skills.
DfES (2007). Raising the attainment of Pakistani, Bangladeshi, Somali and Turkish
heritage pupils (00069-2007BKT-EN). London: Department for Education & Skills.
Greaves, E., Macmillan, L., & Sibieta, L. (2014). Lessons from London schools for
attainment gaps and social mobility. London: Social Mobility and Child Poverty
Commission.
House of Commons (2014). Underachievement in Education by White Working Class
Children:
Education Select Committee’s First Report of Session 2014-15. London: House of
Commons.
Maylor, U., Smart, S., Kuyok, K. A., & Ross, A. (2009). Black Children's Achievement
Programme Evaluation (DCSF-RR177). London: Department for Children, Schools and
Families.
NASUWT (2012). Ethnic minority achievement. Birmingham: National Association of
Schoolteachers and Union on Women Teachers.
OFSTED (1999). Raising the attainment of minority ethnic pupils: School and LEA
responses.
London: Office for Standards in Education.
OFSTED (2002). Achievement of Black Caribbean pupils: Good practice in secondary
schools.
London: Office for Standards in Education.
OFSTED (2004). Managing the Ethnic minority Achievement Grant: Good practice in
primary schools (HMI 2072). London: Office for Standards in Education.
OFSTED (2004). Managing the Ethnic minority Achievement Grant: Good practice in
secondary schools (HMI 2172). London: Office for Standards in Education.
Strand, S. (2012). Disproportionate identification of Black students with special
educational needs ( SEN): Recent national data from England. Paper presented at the
Annual Conference of the British Educational Research Association (BERA), University
of Manchester, 4-6 September 2012.
58
Strand, S. (2013). What accounts for ethnic achievement gaps in secondary schools in
England?
BERA Insights, Issue 4, British Educational Research Association. Available from the
WWW at http://www.bera.ac.uk/publications/Insights [Accessed 7 October 2013).
Strand, S. (2014c). Written evidence submitted to the House of Commons Education
Select Committee inquiry into Underachievement in Education by White Working Class
Children.
Accessed from the WWW 2 October 2014 at:
http://data.parliament.uk/writtenevidence/committeeevidence.svc/evidencedocument/ed
ucation-committee/underachievement-in-education-of-white-working-classchildren/written/2324.pdf
Strand, S. (2014a). Ethnicity, gender, social class and achievement gaps at age 16:
Intersectionality and ‘Getting it’ for the white working class. Research Papers in
Education, 29, (2), 131-171. http://dx.doi.org/10.1080/02671522.2013.767370
Strand, S. (2014b). School effects and ethnic, gender and socio-economic gaps in
educational achievement at age 11. Oxford Review of Education, 40, (2), 223-245.
http://dx.doi.org/10.1080/03054985.2014.891980
Strand, S. (2014c). Mind the gap: An analysis of the FSM gap in Buckinghamshire
County Council. Buckinghamshire: Buckinghamshire County Council. Available from:
https://democracy.buckscc.gov.uk/documents/s49242/Steve%20Strand%20report.pdf
Strand, S., & Demie, F. (2005). English language acquisition and educational
attainment at the end of primary school. Educational Studies, 31, (3), 275-291.
http://dx.doi.org/10.1080/03055690500236613
Strand, S. & Demie, F. (2006). Pupil mobility, attainment and progress in primary
school. British Educational Research Journal, 32, (4), 551-568.
http://dx.doi.org/10.1080/01411920600775191
Strand, S., & Fletcher, J. (2014). A Quantitative Analysis of Exclusions from English
Secondary Schools. University of Oxford: Department for Education.
Strand, S., & Lindsay, G. (2009). Evidence of ethnic disproportionality in special
education in an English population. Journal of Special Education, 43(3), 174-190.
http://dx.doi.org/10.1177/0022466908320461
Strand, S., De Coulon, A., Meschi, E., Vorhaus, J., Ivins, C., Small, L., Sood, A.,
Gervais, M.C. & Rehman, H. (2010). Drivers and challenges in raising the achievement
of pupils from Bangladeshi, Somali and Turkish backgrounds (Research Report DCSFRR226). London: Department for Children School and Families. Access 13 April 2010
59
from
https://www.education.gov.uk/publications/standard/publicationDetail/Page1/DCSFRR226.
Strand, S., Malmberg, L. and Hall, J. (2015). English as an Additional Language (EAL)
and educational achievement: An analysis of the National Pupil Database. London:
Educational Endowment Fund.
http://educationendowmentfoundation.org.uk/uploads/pdf/EAL_and_educational_achiev
ement2.pdf
Whitty, G. & Anders, J. (2014). (How) did New Labour narrow the achievement and
participation gap? London: Centre for Learning and Life Chances in Knowledge
Economies and Societies. http://www.llakes.org
60
Appendices
Appendix 1: DfE Statistical First Releases (SFR) relating to age 16 attainment
Appendix 2: Pupil numbers by ethnic group, entitlement to FSM and gender for 2013
Appendix 3: Notes on Local Authority maps for percentage ethnic minority students
2003-2013
Appendix 4: Percentage of pupils achieving 5AC by ethnic group and gender: 20032013
Appendix 5: GCSE 5EM for pupils not entitled to FSM by ethnic group and gender:
2004-2013
Appendix 6: GCSE 5EM for pupils entitled to FSM by ethnic group and gender: 20042013
Appendix 7: Key stage 2 achievement by ethnic group, gender and entitlement to FSM:
2013
Appendix 8: Foundation Stage Profile achievement by ethnic group, gender and
entitlement to FSM: 2013
Appendix 9: Special educational needs (SEN) and fixed term exclusions (FTE) by
ethnic group.
61
Appendix 1: DfE Statistical First Releases (SFR) relating to
age 16 attainment
Exam
results
year SFR ID
2003 SFR 2004-04
Title
National Curriculum Assessment and GCSE/GNVQ Attainment by
Pupil Characteristics in England 2002 (final) and 2003 (provisional)
National Curriculum Assessment, GCSE and Equivalent Attainment
and Post-16 Attainment by Pupil Characteristics in England 2004
National Curriculum Assessment, GCSE and Equivalent Attainment
and Post-16 Attainment by Pupil Characteristics in England 2005
National Curriculum Assessment, GCSE and Equivalent Attainment
and Post-16 Attainment by Pupil Characteristics in England 2005/06
National Curriculum Assessment, GCSE and Equivalent Attainment
and Post-16 Attainment by Pupil Characteristics in England 2006/07
2004
SFR 2005-08
2005
SFR 2006-09
2006
SFR 2007-04
2007
SFR 2007-38
2008
SFR 2008-32
Attainment by pupil characteristics in England, 2007/08
2009
SFR 2009-34
GCSE Attainment by Pupil Characteristics, in England 2008/09
2010
SFR 2010-37
2011
SFR 2012-03
2012
SFR 2013-04
2013
SFR 2014-05
GCSE and equivalent attainment by Pupil Characteristics, in England
2009/10
GCSE and equivalent attainment by Pupil Characteristics, in England
2010/11
GCSE and equivalent attainment by Pupil Characteristics, in England
2011/12
GCSE and equivalent attainment by Pupil Characteristics, in England
2012/13
62
Appendix 2: Pupil numbers at the end of key stage 4 by ethnic group, entitlement to FSM and
gender for 2013
Boys
Eligible for FSM
Girls
Total
Boys
All Other Pupils
Girls
Total
Boys
All pupils
Girls
Total
White
White British
Irish
Traveller of Irish heritage
Gypsy / Roma
White other background
Mixed
White and Black Caribbean
White and Black African
White and Asian
Other mixed background
Asian
Indian
Pakistani
Bangladeshi
Other Asian background
Black
Black Caribbean
Black African
Black other background
Chinese
Any other ethnic group
29,790
27,977
168
42
191
1,412
2,290
915
249
362
764
5,313
683
2,588
1,439
603
4,204
1,059
2,675
470
85
1,161
28,636
26,923
120
43
201
1,349
2,270
913
288
343
726
5,142
625
2,388
1,520
609
4,218
1,000
2,764
454
83
1,100
58,426
54,900
288
85
392
2,761
4,560
1,828
537
705
1,490
10,455
1,308
4,976
2,959
1,212
8,422
2,059
5,439
924
168
2,261
205,023
195,617
850
27
197
8,332
8,480
2,676
909
1,901
2,994
18,636
6,262
6,635
2,355
3,384
9,428
3,004
5,336
1,088
1,060
2,640
197,141
187,952
761
25
231
8,172
8,571
2,829
885
1,828
3,029
17,695
5,973
6,167
2,362
3,193
9,592
3,095
5,426
1,071
1,029
2,343
402,164
383,569
1,611
52
428
16,504
17,051
5,505
1,794
3,729
6,023
36,331
12,235
12,802
4,717
6,577
19,020
6,099
10,762
2,159
2,089
4,983
234,813
223,594
1,018
69
388
9,744
10,770
3,591
1,158
2,263
3,758
23,949
6,945
9,223
3,794
3,987
13,632
4,063
8,011
1,558
1,145
3,801
225,777
214,875
881
68
432
9,521
10,841
3,742
1,173
2,171
3,755
22,837
6,598
8,555
3,882
3,802
13,810
4,095
8,190
1,525
1,112
3,443
460,590
438,469
1,899
137
820
19,265
21,611
7,333
2,331
4,434
7,513
46,786
13,543
17,778
7,676
7,789
27,442
8,158
16,201
3,083
2,257
7,244
All pupils
43,330
41,852
85,182
247,678
238,474
486,152
291,008
280,326
571,334
Source: DfE SFR 2014-05.
63
Appendix 3: Notes on the Local Authority maps for
percentage ethnic minority students 2003 & 2013
The percentage categories that are shown in the Local Authority (LA) maps of the report
are used to best represent the diversity of percentage values for the different LAs. The
relatively extreme percentage values for the LAs – many having relatively small
proportions of ethnic minority students and many having relatively high proportions of
ethnic minority students – meant that determining percentage categories based on
quintiles gave percentage categories of very different widths. Likewise, using categories
that had equal widths (in terms of percentage) obscured the large diversity in the data.
Therefore, the percentage categories were imputed using the Jenks natural breaks
classification method. This method imputes category boundaries such that values in the
category have a minimal standard deviation from the category’s mean and a maximum
deviation from the other categories. These category boundaries were then rounded to
make more meaningful categories.
Number of LAs by category 2003 and 2013
Percentage
categories
(%)
0 - 6.99
7 - 15.99
16 - 30.99
31 - 49.99
50+
Total
Year
2003
2013
61
36
20
11
24
152
28
38
35
18
33
152
Two LAs were reorganised in 2009. The county of Bedfordshire was abolished and split
into two LAs (Bedford & Central Bedfordshire) and the county of Cheshire was also
abolished and split into two LAs (Cheshire East and Cheshire West and Chester). The
new LAs, particularly Bedford & Central Bedfordshire, have very divergent ethnic
profiles from each other and this is reflected in the 2013 data. However, for 2003 it is
not possible to disaggregate data for the new LAs so the original LA average has to be
used for both new LAs in each case.
64
Appendix 4: GCSE 5AC by ethnic group and gender: 2003-2013
Gender
Boys
Boys
Boys
Boys
Boys
Boys
Boys
Boys
Boys
Boys
Boys
Ethnic group
White British
White Other groups
Mixed White & Black Carib.
Indian
Pakistani
Bangladeshi
Black Caribbean
Black African
Chinese
Black other groups
ALL BOYS
2003
46.1
46.3
32.3
60.3
35.7
38.5
25.1
34.1
70.9
27.2
45.5
2004
47.4
49.3
34.1
61.6
38.8
41.0
27.3
37.3
69.5
29.8
46.8
2005
50.2
54.2
37.6
64.8
43.2
46.7
33.3
42.9
77.1
33.7
49.9
2006
52.6
54.4
39.3
66.7
44.9
50.3
35.9
44.3
74.6
38.9
52.2
2007
55.2
54.8
42.0
69.7
47.0
52.3
41.5
50.0
81.6
42.9
54.8
2008
59.5
56.6
49.7
74.3
52.7
56.0
46.9
53.5
80.9
49.9
59.1
2009
65.9
63.5
58.6
78.8
61.2
65.5
56.4
65.7
84.1
60.0
65.8
2010
71.8
70.6
66.1
85.0
69.8
72.0
64.2
71.6
87.6
65.9
71.9
2011
76.8
76.9
70.8
87.6
77.4
79.4
72.2
79.1
90.6
73.9
77.0
2012
79.6
76.9
75.1
89.3
79.9
82.9
75.3
80.5
91.9
76.6
79.8
2013
79.2
77.9
76.7
88.6
81.0
81.5
76.3
81.3
90.9
79.3
79.6
Girls
Girls
Girls
Girls
Girls
Girls
Girls
Girls
Girls
Girls
Girls
White British
White Other groups
Mixed White & Black Carib.
Indian
Pakistani
Bangladeshi
Black Caribbean
Black African
Chinese
Black Other groups
ALL GIRLS
2003
56.6
58.2
46.8
70.3
48.1
52.6
40.3
46.8
79.2
40.3
56.1
2004
57.4
61.6
44.9
71.9
52.1
55.2
43.8
48.9
79.1
43.0
57.0
2005
60.1
63.7
50.0
75.8
54.1
58.5
49.4
53.3
85.1
50.8
60.0
2006
61.9
64.8
53.8
76.5
57.6
61.8
52.4
56.1
84.3
55.7
61.9
2007
64.0
63.1
55.8
79.4
59.6
64.4
56.2
61.1
85.1
57.3
63.9
2008
68.2
65.1
60.6
82.7
64.0
68.9
60.8
66.8
87.6
62.5
68.2
2009
73.8
70.4
67.9
85.8
72.0
73.8
69.9
74.1
91.2
68.4
73.9
2010
79.3
77.1
75.5
89.7
78.4
79.9
76.2
80.5
92.3
77.3
79.5
2011
83.9
81.2
80.1
92.8
84.1
86.2
82.6
84.7
95.0
84.1
84.0
2012
86.3
83.8
82.7
93.1
85.6
87.6
83.9
86.7
94.2
82.8
86.3
2013
86.1
83.6
84.0
93.6
86.4
88.4
84.5
87.1
95.2
85.6
86.5
All
All
All
All
All
All
All
All
All
All
All
White British
White Other groups
Mixed White & Black Carib.
Indian
Pakistani
Bangladeshi
Black Caribbean
Black African
Chinese
Black Other groups
ALL PUPILS
2003
51.3
52.0
39.9
65.2
41.5
45.5
32.9
40.7
74.8
33.6
50.7
2004
52.3
55.0
39.7
66.6
45.2
48.4
35.7
43.3
74.2
36.2
51.9
2005
55.0
58.9
44.1
70.1
48.4
52.7
41.7
48.3
81.0
41.7
54.9
2006
57.2
59.6
46.9
71.4
50.9
56.2
44.4
50.3
79.3
46.9
56.9
2007
59.5
58.8
48.8
74.4
53.0
58.4
49.1
55.6
83.3
49.7
59.3
2008
63.8
60.8
55.3
78.3
58.2
62.3
54.0
60.3
84.3
56.2
63.5
2009
69.8
66.9
63.4
82.3
66.4
69.7
63.3
70.0
87.5
64.2
69.8
2010
75.5
73.8
70.9
87.3
74.0
75.9
70.3
76.2
89.9
71.4
75.6
2011
80.2
78.5
75.6
90.1
80.5
82.8
77.5
82.0
92.7
78.9
80.5
2012
82.9
80.3
79.0
91.1
82.7
85.3
79.7
83.8
93.1
79.6
83.0
2013
82.7
80.7
80.4
91.0
83.6
85.0
80.4
84.2
93.0
82.4
82.9
All
All
All
All
All
All
All
All
All
All
Odds ratios
White British
White Other groups
Mixed White & Black Carib.
Indian
Pakistani
Bangladeshi
Black Caribbean
Black African
Chinese
Black Other groups
2003
1.03
0.63
1.78
0.67
0.79
0.47
0.65
2.82
0.48
2004
1.11
0.60
1.82
0.75
0.86
0.51
0.70
2.62
0.52
2005
1.17
0.65
1.92
0.77
0.91
0.59
0.76
3.49
0.59
2006
1.10
0.66
1.87
0.78
0.96
0.60
0.76
2.87
0.66
2007
0.97
0.65
1.98
0.77
0.96
0.66
0.85
3.40
0.67
2008
0.88
0.70
2.05
0.79
0.94
0.67
0.86
3.05
0.73
2009
0.87
0.75
2.01
0.86
1.00
0.75
1.01
3.03
0.78
2010
0.91
0.79
2.23
0.92
1.02
0.77
1.04
2.89
0.81
2011
0.90
0.76
2.25
1.02
1.19
0.85
1.12
3.14
0.92
2012
0.84
0.78
2.11
0.99
1.20
0.81
1.07
2.78
0.80
2013
0.87
0.86
2.12
1.07
1.19
0.86
1.11
2.78
0.98
65
Appendix 5: GCSE 5EM for pupils not entitled to FSM by ethnic group and gender: 2004-2013
Ethnic Group
2004
2005
2006
2007
Boys
2008 2009
2010
2011
2012
2013
2004
2005
2006
2007
2008
Girls
2009 2010
2011
2012
2013
-
42.4
43.7
45.4
47.8
50.7
54.8
58.3
57.6
59.1
-
51.0
52.2
53.6
55.7
57.7
62.2
65.4
67.3
69.5
White British
40.9
42.2
43.6
45.3
47.8
50.8
55.0
58.5
57.9
59.4
49.1
50.8
52.1
53.6
55.8
57.8
62.4
65.6
67.7
69.8
Irish
47.4
51.6
52.7
53.9
57.9
62.9
66.4
70.3
67.7
69.9
56.9
62.2
60.0
62.2
67.0
65.3
71.7
73.1
77.4
79.1
Traveller of Irish Heritage
32.7
33.3
x
x
x
x
33.3
27.3
23.5
14.8
47.6
30.6
x
20.0
x
18.9
22.6
21.7
23.7
36.0
Gypsy / Roma
18.4
9.2
x
6.3
4.6
6.3
5.9
11.7
7.8
13.2
12.1
13.3
x
10.4
10.6
15.2
13.9
12.2
12.3
22.1
Any other White background
45.1
47.0
47.1
46.3
45.1
46.3
49.0
52.9
49.9
52.8
53.6
55.7
56.0
54.0
53.2
54.6
56.7
59.9
59.6
62.0
-
42.8
44.1
45.4
48.6
53.3
55.7
59.4
60.3
62.7
-
52.4
52.8
54.1
55.6
59.3
63.3
67.1
69.1
72.3
White and Black Caribbean
29.3
29.8
31.7
32.9
38.6
42.9
45.8
48.8
52.5
54.8
35.8
40.2
42.8
44.6
46.8
50.5
54.0
58.8
63.4
66.3
White and Black African
37.0
41.6
43.4
46.7
49.3
50.8
58.2
58.5
58.8
61.4
46.5
51.6
53.5
47.9
55.3
62.2
62.7
65.1
69.3
74.2
White and Asian
59.3
58.5
60.4
61.3
60.4
65.2
65.5
68.0
69.9
70.5
66.2
64.6
67.9
67.6
65.9
68.8
73.9
77.0
74.3
77.2
Any other mixed background
43.1
46.1
47.2
48.1
51.6
57.1
58.3
63.7
61.6
65.3
51.7
57.5
54.7
58.1
59.0
62.3
66.3
69.6
71.2
74.4
-
46.3
46.9
48.0
50.2
52.4
57.6
61.5
61.4
62.4
-
55.2
56.7
59.3
60.9
62.9
66.8
70.2
70.7
72.8
Indian
52.9
55.2
57.0
59.3
62.4
64.8
69.6
72.5
72.3
72.0
62.0
65.7
66.7
69.7
72.3
73.7
77.0
80.3
80.6
82.7
Pakistani
32.4
34.5
34.8
36.5
38.9
41.9
49.3
53.1
53.1
54.2
42.3
41.6
43.3
46.8
48.8
52.4
56.8
60.9
62.2
63.8
Bangladeshi
33.4
35.9
39.8
41.8
41.7
48.7
51.6
59.2
60.6
62.3
41.2
43.9
46.9
51.7
54.6
57.0
61.6
65.9
68.5
71.7
Any other Asian background
48.9
51.6
48.4
46.4
48.8
48.7
52.6
57.8
57.7
60.7
59.1
60.9
64.7
62.7
62.2
63.8
69.0
70.8
70.6
72.5
-
29.6
31.6
35.0
37.9
42.3
46.6
52.6
53.2
57.2
-
41.3
44.8
47.4
51.7
55.5
59.5
64.8
64.2
67.7
Black Caribbean
20.3
24.1
25.2
29.7
31.8
35.8
39.8
45.1
46.4
50.4
32.2
35.2
39.8
41.8
46.3
48.4
52.9
58.8
59.1
63.4
Black African
33.2
36.6
38.5
40.4
43.0
47.9
52.4
57.6
58.1
61.5
45.1
48.2
50.7
53.5
56.4
61.9
64.9
69.4
67.8
70.9
Any other Black background
23.7
26.6
29.3
31.7
36.5
38.9
43.8
52.9
48.1
54.8
33.1
40.3
41.5
43.0
49.4
49.7
56.5
61.5
60.5
64.4
Chinese
58.1
64.7
61.2
68.1
64.9
64.8
70.6
74.6
72.6
74.2
72.2
75.4
73.8
75.4
76.6
79.3
80.9
83.5
81.4
82.4
40.9
42.7
43.5
41.9
43.6
44.8
50.3
51.5
52.7
59.3
48.8
51.9
53.4
52.7
55.7
56.1
59.4
62.5
63.8
66.6
White
Mixed
Asian
Black
Any other ethnic group
All pupils
40.6
42.1
43.4
45.2
47.6
50.6
54.7
58.3
57.8
59.5
48.9
50.9
52.2
53.7
55.9
58.1
62.4
65.8
67.5
69.8
Odd-Ratio
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
White Other groups
1.19
1.21
1.15
1.04
0.90
0.84
0.79
0.80
0.72
0.76
1.20
1.22
1.17
1.02
0.90
0.88
0.79
0.78
0.70
0.71
Mixed White & Black Caribbean
0.60
0.58
0.60
0.59
0.69
0.73
0.69
0.68
0.80
0.83
0.58
0.65
0.69
0.70
0.70
0.74
0.71
0.75
0.83
0.85
Indian
1.62
1.69
1.71
1.76
1.81
1.78
1.87
1.87
1.90
1.76
1.69
1.86
1.84
1.99
2.07
2.05
2.02
2.14
1.98
2.07
Pakistani
0.69
0.72
0.69
0.69
0.70
0.70
0.80
0.80
0.82
0.81
0.76
0.69
0.70
0.76
0.75
0.80
0.79
0.82
0.79
0.76
Bangladeshi
0.72
0.77
0.86
0.87
0.78
0.92
0.87
1.03
1.12
1.13
0.73
0.76
0.81
0.93
0.95
0.97
0.97
1.01
1.04
1.10
Black African
0.72
0.79
0.81
0.82
0.82
0.89
0.90
0.96
1.01
1.09
0.85
0.90
0.95
1.00
1.02
1.19
1.11
1.19
1.00
1.05
Black Caribbean
0.37
0.43
0.44
0.51
0.51
0.54
0.54
0.58
0.63
0.69
0.49
0.53
0.61
0.62
0.68
0.68
0.68
0.75
0.69
0.75
Chinese
2.00
2.51
2.04
2.58
2.02
1.78
1.96
2.08
1.93
1.97
2.69
2.97
2.59
2.65
2.59
2.80
2.55
2.65
2.09
2.03
66
Appendix 6: GCSE 5EM for pupils entitled to FSM by ethnic group and gender: 2004-2013
Boys
Girls
Ethnic group
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
White
12.7
14.0
15.2
16.6
19.4
23.3
26.5
26.9
28.3
24.3
-
17.5
18.5
20.3
22.6
24.5
28.4
32.2
35.1
37.1
White British
12.2
12.3
13.8
14.8
16.3
19.0
22.8
26.0
26.4
27.9
23.8
15.9
17.1
18.2
20.0
22.0
24.0
27.9
31.7
34.7
36.8
Irish
17.8
22.1
16.5
22.8
17.8
22.4
28.8
35.3
29.2
38.7
21.9
21.4
20.8
22.4
25.3
30.8
26.0
29.5
31.7
36.4
38.3
x
0.0
x
0.0
0.0
x
13.3
11.8
17.6
11.9
13.2
5.3
3.7
x
x
x
x
19.4
12.8
7.0
14.0
2.6
3.3
x
5.3
6.3
4.4
5.5
5.3
7.4
9.9
3.0
1.4
8.6
x
x
4.5
9.2
7.1
11.2
9.5
8.5
17.5
20.1
20.9
23.8
24.2
29.1
34.5
37.8
37.7
39.1
36.3
23.2
27.3
26.7
29.8
34.5
34.6
39.5
45.6
46.3
48.6
17.6
18.5
18.1
24.3
26.6
31.7
36.1
36.3
39.5
32.8
-
24.5
25.3
30.8
31.2
34.7
38.3
42.8
46.3
48.2
White and Black Caribbean
11.2
14.8
15.2
13.5
19.5
23.7
28.4
30.8
30.4
34.6
27.2
16.2
20.6
20.7
25.1
26.5
30.9
31.7
36.4
41.0
40.4
White and Black African
21.5
15.6
20.8
20.0
28.2
26.2
33.7
35.7
37.8
44.2
34.9
27.1
31.0
29.3
34.1
28.0
31.5
42.2
50.2
51.9
52.4
White and Asian
27.1
22.6
27.8
25.3
32.2
29.2
34.9
38.5
40.5
38.7
37.7
28.1
32.2
34.4
33.6
36.4
43.3
42.2
46.7
52.3
57.7
Any other mixed background
16.5
20.1
18.9
20.8
25.9
29.3
34.1
41.5
40.6
44.2
36.4
20.3
24.6
26.4
36.6
36.6
37.4
43.7
46.4
49.2
51.9
24.9
27.6
29.2
32.7
35.1
41.5
45.7
48.3
48.6
44.0
-
32.6
36.8
38.5
42.8
44.3
50.2
53.4
55.2
57.2
Indian
29.6
32.4
33.9
36.2
41.3
43.3
51.0
51.3
54.9
56.7
50.1
41.2
42.9
45.7
48.3
50.5
52.6
59.5
63.0
60.7
66.7
Pakistani
20.1
20.2
23.3
26.1
28.1
30.4
36.6
39.3
43.4
42.8
39.0
25.1
28.4
31.4
33.1
37.0
38.5
44.8
46.9
49.6
51.1
Bangladeshi
25.6
27.4
31.6
31.6
35.9
37.7
46.8
53.6
55.7
55.5
51.5
32.7
33.4
38.8
41.0
47.6
47.8
54.0
58.7
61.4
62.8
Any other Asian background
23.2
30.4
27.2
28.3
35.6
41.9
37.8
48.1
43.7
47.8
41.4
37.2
35.7
43.9
45.0
46.5
51.2
54.4
56.0
56.5
57.0
9.7
16.1
17.9
22.5
25.8
28.8
33.6
39.5
40.3
43.1
37.3
-
24.6
28.3
31.6
34.7
38.5
44.6
48.9
50.7
53.3
13.6
14.9
17.8
23.3
22.4
27.6
33.2
32.1
36.9
30.9
18.1
24.2
24.0
30.2
29.0
36.0
38.7
42.1
47.7
47.7
Black African
16.0
18.7
20.4
25.8
27.0
31.7
36.1
42.4
43.8
46.5
40.8
22.2
26.0
30.5
32.4
37.0
39.4
47.8
51.8
52.9
56.0
Any other Black background
14.6
11.4
13.7
16.2
24.8
26.3
32.9
36.0
36.8
37.7
30.7
17.9
18.4
28.7
31.2
35.9
39.1
37.3
47.0
43.4
48.7
55.1
45.9
47.2
53.4
55.3
64.2
60.6
61.4
62.4
74.1
60.5
55.8
72.7
60.0
67.6
71.5
77.7
76.4
86.1
73.6
79.5
19.9
23.2
26.1
27.3
31.0
35.8
39.3
46.0
46.7
46.5
48.0
29.4
32.1
33.9
40.4
40.7
49.1
49.6
51.1
56.1
56.8
Traveller of Irish Heritage
Gypsy / Roma
Any other White background
Mixed
Asian
Black
Black Caribbean
Chinese
Any other ethnic group
All pupils
14.6
15.2
16.8
18.2
20.5
23.4
27.8
31.4
32.0
33.5
29.2
19.0
20.8
22.5
24.7
27.3
29.8
34.2
37.9
40.6
42.5
Odd-Ratio
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
White Other groups
1.53
1.79
1.65
1.80
1.64
1.75
1.78
1.73
1.69
1.66
1.82
1.60
1.82
1.64
1.70
1.87
1.68
1.69
1.81
1.62
1.62
Mixed White & Caribbean
0.91
1.24
1.12
0.90
1.24
1.32
1.34
1.27
1.22
1.37
1.20
1.02
1.26
1.17
1.34
1.28
1.42
1.20
1.23
1.31
1.16
Indian
3.03
3.42
3.20
3.27
3.61
3.26
3.52
3.00
3.39
3.38
3.21
3.71
3.64
3.78
3.74
3.62
3.51
3.80
3.67
2.91
3.44
Pakistani
1.81
1.80
1.90
2.03
2.01
1.86
1.95
1.84
2.14
1.93
2.05
1.77
1.92
2.06
1.98
2.08
1.98
2.10
1.90
1.85
1.79
Bangladeshi
2.48
2.69
2.89
2.66
2.88
2.58
2.98
3.29
3.51
3.22
3.40
2.57
2.43
2.85
2.78
3.22
2.90
3.03
3.06
2.99
2.90
Black African
1.37
1.64
1.60
2.00
1.90
1.98
1.91
2.10
2.17
2.25
2.21
1.51
1.70
1.97
1.92
2.08
2.06
2.37
2.32
2.11
2.19
Black Caribbean
0.77
1.12
1.09
1.25
1.56
1.23
1.29
1.41
1.32
1.51
1.43
1.17
1.55
1.42
1.73
1.45
1.78
1.63
1.57
1.72
1.57
Chinese
8.83
6.05
5.58
6.60
6.35
7.65
5.21
4.53
4.63
7.39
4.90
6.68
12.91
6.74
8.35
8.89
11.03
8.37
13.35
5.25
6.66
67
Appendix 7: Key stage 2 achievement by ethnic group, gender and entitlement to FSM: 2013
Key Stage 2 Reading,
writing & mathematics
Pupils known to be eligible for FSM
Number of eligible pupils
All pupils
3
Boys
Girls
Total
All other pupils
Percentage
achieving level 4 or
above
Boys
Girls
Total
2
All pupils
Percentage
achieving level 4 or
above
Number of eligible pupils
Boys
Girls
Total
Boys
Girls
Total
Number of eligible pupils
Boys
Girls
Total
2
Percentage
achieving level 4 or
above
Boys
Girls
Total
50,036
47,713
97,749
56
65
60
222,787
212,579
435,366
76
82
79
272,823
260,292
533,115
72
79
75
White
34,313
32,287
66,600
52
62
57
177,352
168,718
346,070
76
83
79
211,665
201,005
412,670
72
79
76
white British
31,882
29,966
61,848
53
63
58
166,211
157,980
324,191
77
83
80
198,093
187,946
386,039
73
80
76
Irish
157
160
317
54
66
60
712
717
1,429
84
88
86
869
877
1,746
79
84
82
traveller of Irish heritage
161
122
283
29
39
33
54
45
99
26
53
38
215
167
382
28
43
35
Gypsy / Roma
409
370
779
17
19
18
378
365
743
24
31
28
787
735
1,522
21
25
23
any other white background
1,704
1,669
3,373
53
62
57
9,997
9,611
19,608
66
73
70
11,701
11,280
22,981
64
72
68
Mixed
3,166
3,132
6,298
61
69
65
8,923
8,560
17,483
78
84
81
12,089
11,692
23,781
73
80
77
white and black Caribbean
1,238
1,216
2,454
58
68
63
2,456
2,387
4,843
73
80
76
3,694
3,603
7,297
68
76
72
white and black African
392
361
753
60
72
66
1,020
986
2,006
75
82
78
1,412
1,347
2,759
71
79
75
white and Asian
481
483
964
60
69
64
2,149
2,045
4,194
82
87
84
2,630
2,528
5,158
78
83
81
any other mixed background
1,055
1,072
2,127
65
70
67
3,298
3,142
6,440
79
85
82
4,353
4,214
8,567
76
81
79
Asian
5,785
5,590
11,375
66
70
68
22,042
21,091
43,133
75
81
78
27,827
26,681
54,508
73
79
76
Indian
633
626
1,259
69
74
72
6,351
6,007
12,358
82
86
84
6,984
6,633
13,617
81
85
83
Pakistani
2,946
2,801
5,747
63
68
65
8,977
8,621
17,598
70
76
73
11,923
11,422
23,345
68
74
71
Bangladeshi
1,525
1,522
3,047
71
73
72
3,284
3,192
6,476
76
81
78
4,809
4,714
9,523
74
78
76
681
641
1,322
64
69
66
3,430
3,271
6,701
77
84
80
4,111
3,912
8,023
75
81
78
Black
5,091
5,110
10,201
63
70
66
9,101
8,838
17,939
72
81
77
14,192
13,948
28,140
69
77
73
black Caribbean
1,285
1,303
2,588
57
67
62
2,456
2,392
4,848
68
79
74
3,741
3,695
7,436
64
75
70
black African
3,202
3,231
6,433
65
72
69
5,457
5,367
10,824
75
82
79
8,659
8,598
17,257
71
79
75
604
576
1,180
61
66
63
1,188
1,079
2,267
69
80
74
1,792
1,655
3,447
66
75
70
72
72
144
86
88
87
773
874
1,647
81
88
85
845
946
1,791
82
88
85
1,254
1,215
2,469
63
68
65
2,789
2,678
5,467
70
75
73
4,043
3,893
7,936
67
73
70
any other Asian background
any other black background
Chinese
any other ethnic group
Source: National Pupil Database
1. Figures for 2012 are based on final data, 2013 figures are based on revised data.
2. Includes pupils not eligible for free school meals and for whom free school meal eligibility was unclassified or could not be determined.
3. Includes pupils for whom ethnicity was not obtained, refused or could not be determined.
68
Appendix 8: Foundation Stage Profile achievement by ethnic group, gender and entitlement to FSM
A good level of
development
All pupils
8
White
white British
6
Pupils known to be eligible for free school meals
% achieving a good
2
Number of eligible pupils
level of development
Boys
Girls
Total Boys Girls Total
All other pupils
6
Number of eligible pupils
2
Boys
Girls
Total
4
All pupils
% achieving a good
level of development
Boys Girls Total
6
Number of eligible pupils
2
Boys
Girls
Total
% achieving a good
level of development
Boys Girls Total
60,642
58,303
118,945
29
44
36
268,562
255,795
524,357
47
63
55
329,204
314,098
643,302
44
60
52
39,473
38,112
77,585
27
42
34
191,467
181,379
372,846
49
66
57
230,940
219,491
450,431
45
62
53
36,995
35,747
72,742
27
43
35
174,630
165,215
339,845
50
67
58
211,625
200,962
412,587
46
63
54
Irish
152
146
298
28
48
38
696
687
1,383
55
69
62
848
833
1,681
50
65
58
traveller of Irish heritage
139
159
298
12
16
14
123
121
244
29
40
35
262
280
542
20
27
24
337
314
651
9
17
12
393
399
792
14
26
20
730
713
1,443
11
22
16
1,850
1,746
3,596
27
42
34
15,625
14,957
30,582
36
49
42
17,475
16,703
34,178
35
48
41
4,646
4,575
9,221
32
50
41
12,470
12,117
24,587
50
66
58
17,116
16,692
33,808
45
62
53
1,747
1,661
3,408
31
48
39
2,807
2,704
5,511
47
65
55
4,554
4,365
8,919
41
59
49
699
735
1,434
29
52
41
1,693
1,710
3,403
48
65
57
2,392
2,445
4,837
43
61
52
3,330
3,217
6,547
53
69
61
4,044
3,921
7,965
50
65
57
Gypsy / Roma
any other white
background
Mixed
white and black
Caribbean
white and black African
white and Asian
any other mixed
background
714
704
1,418
33
48
41
1,486
1,475
2,961
33
51
42
4,640
4,486
9,126
50
66
58
6,126
5,961
12,087
46
62
54
Asian
4,489
4,327
8,816
30
44
37
26,408
25,332
51,740
42
56
49
30,897
29,659
60,556
41
54
47
504
444
948
36
47
41
7,925
7,681
15,606
51
64
58
8,429
8,125
16,554
50
64
57
1,974
1,861
3,835
27
41
34
9,821
9,414
19,235
36
49
42
11,795
11,275
23,070
35
48
41
Bangladeshi
any other Asian
background
1,067
1,059
2,126
33
49
41
3,548
3,334
6,882
38
54
46
4,615
4,393
9,008
37
53
45
944
963
1,907
30
46
38
5,114
4,903
10,017
43
57
50
6,058
5,866
11,924
41
55
48
Black
5,727
5,516
11,243
37
52
44
10,611
10,451
21,062
46
62
54
16,338
15,967
32,305
43
59
51
1,166
1,194
2,360
32
53
43
1,943
1,935
3,878
45
61
53
3,109
3,129
6,238
40
58
49
3,788
3,591
7,379
38
52
45
7,308
7,175
14,483
47
63
55
11,096
10,766
21,862
44
59
51
773
731
1,504
35
53
44
1,360
1,341
2,701
45
60
53
2,133
2,072
4,205
42
58
49
122
120
242
32
42
37
1,159
1,129
2,288
45
56
50
1,281
1,249
2,530
43
55
49
Indian
Pakistani
black Caribbean
black African
any other black
background
Chinese
any other ethnic group
1,313
1,211
2,524
32
45
38
3,735
1. Figures based on final data.
2. Only includes pupils with a valid result for every achievement scale.
3. All English providers of state-funded early years education (including academies and free schools),
private, voluntary and independent (PVI) sectors are within the scope of the EYFSP data collection.
Data for any children in the PVI sector no longer in receipt of funding who were included in the return
submitted by the LA to DfE will not be included in the figures. See technical notes in the accompanying
SFR text for further information.
4. Includes pupils not eligible for free school meals and for whom free school meal eligibility was
unclassified or could not be determined.
3,531
7,266
39
53
46
5,048
4,742
9,790
37
51
44
5. Achieved at least the expected standard all areas of learning (proportion achieving ‘expected’ or
‘exceeded’ in all 17 Early Learning Goals(ELGs))
6. A pupil achieving at least the expected level in the ELGs within the three prime areas of learning and
within literacy and numeracy is classed as having "a good level of development".
7. Average point score for each characteristic grouping. This is a supporting measure taking into
account performance across all 17 ELGs, 1 point for emerging, 2 for expected and 3 for exceeding.
The sum is then taken for all pupils with that characteristic and the mean given.
8. Includes pupils for whom ethnicity was not obtained, was refused or could not be determined.
69
Appendix 9: Special educational needs (SEN) and fixed term
exclusions (FTE)
Special educational needs (SEN)
There are wide variations in the proportion of students from different ethnic groups
identified with various special educational needs (SEN). Overall 10% of students are
identified with SEN at School Action Plus or with a formal statement so the numbers are
not small (Strand, 2012). Strand & Lindsay (2009) analysed the 2005 national data for
over 6.4M students aged 5-16 in England revealing that Black Caribbean and Mixed
White & Black Caribbean students are twice as likely to be identified with behavioural,
emotional and social difficulties (BESD) as White British students, and still 1.5 times
more likely after adjusting for economic disadvantage. Strand (2012) updates the
analysis for 2007, 2009 and 2011 but reports the over-representation has not decreased
in subsequent years. Figure 19 shows the analysis for 2011.
Strand & Lindsay (2012) also identify Chinese and Black Caribbean students are overidentified in relation to speech language and communication needs (SLCN) even after
control for SES, and all Asian groups were substantially under-represented relative to
White British students for autistic spectrum disorders (ASD) indicating potential
undiagnosed need.
Figure 21: Unadjusted and adjusted ethnic group Odds Ratios for identification of BESD: January
2011
Source: Date for all students in England aged 5-16 as at January 2011. Outcome is all students at
School Action Plus or statemented with a primary need of behavioural, emotional or social difficulties
(BESD). Adjusted ORs taken account of age, gender, entitlement to FSM, IDACI. See Strand (2012)
for details.
Exclusions from school
Strand and Fletcher (2014) tracked and analysed the fixed term exclusions for an
England national cohort over 560,000 students between the age of 11 in 2007 until they
were age 16 in 2011. The number of students experiencing one or more fixed-term
exclusion was not small, with around 16% of students experiencing one or more fixed
term exclusions during their secondary school career. However, the figure was
substantially higher for Mixed White & Black Caribbean (31%) and for Black Caribbean
(33%) students, while just 8% for Indian and 4% for Chinese students. The results are
presented in Figure 20.
Again the over-representation of Black Caribbean and Mixed White & Black Caribbean
students persisted (OR=1.75) compared to White British students even after adjusting
for a range of pupil background covariates such as age, gender, poverty (FSM and
IDACI), KS2 (age 11) test scores and level of attendance during the first term of Y7.
71
Figure 22: Number of exclusions during secondary school (Y7-Y11)
Source: Data drawn from a longitudinal sample of 550,000 students in England followed between the age
of 11 (in 2007) and 16 (in 2011). See Strand & Fletcher (2014) for full details.
72
© National Institute of Economic and Social Research & University of Oxford 2015
Reference: DFE-RR439B
ISBN: 978-1-78105-499-4
This research was commissioned under the 2010 to 2015 Conservative and Liberal
Democrat coalition government. Views expressed in this report are those of the authors.
They do not necessarily reflect government policy.
The views expressed in this report are the authors’ and do not necessarily reflect those
of the Department for Education.
Any enquiries regarding this publication should be sent to us at:
Nicola.Mackenzie@education.gsi.gov.uk or www.education.gov.uk/contactus
This document is available for download at www.gov.uk/government/publications
73